With a Grain of Salt: Uncertain News and Firms Disclosures

Key Question Being Explored in This Insight:

Beatrice Michaeli discusses how uncertain veracity of external news influences investor beliefs, market prices and corporate disclosures.

View the page of: Beatrice Michaeli

Institution: University of California Los Angeles

Explore more insights in:

Accounting Professional

Media Literacy Toolkit

Faculti's Media Literacy Toolkit helps viewers critically engage with academic insights by analyzing the research context, identifying perspectives, and encouraging thoughtful evaluation.

Critical Questions to Consider ▼

What assumptions does the research make?
Are there alternative perspectives not explored?
What are the limitations of the research method?

Bias and Perspective Awareness ▼

This research comes from University of California Los Angeles . Reflect on how the institution's academic focus and research partnerships may shape the questions being explored.

Further Reading ▼

Read the Study

Beatrice Michaeli discusses how uncertain veracity of external news influences investor beliefs, market prices and corporate disclosures.

Image courtesy of interviewee. November 22, 2024

Log-in or Sign-up to Faculti
Currently viewing this subject insight as a guest. You have insight(s) remaining for this month. Login to view 8000+ figures on the platform.

Generating Workspace...

Export as PDF

This video discusses the impacts of misinformation and external news on capital markets, using historical examples from incidents involving United Airlines and Eli Lilly. It explains how false information, often propagated through social media, can lead to market chaos, regardless of the truth behind the claims. The narrative emphasizes the role of outside news, including analyst forecasts and media reports, in shaping investor beliefs and reactions.

The authors present a model to analyze how firms respond to this outside information, particularly in the context of disclosing private information to the market. They note that investors often perceive bad news as more credible than good news, which can lead to unexpected market reactions—specifically, a non-monotonic response where better news may actually result in a decrease in stock prices.

This video concludes by highlighting the need for empirical researchers to consider the uncertain informativeness of external news in their analyses, suggesting that understanding these dynamics will improve insights into market behavior and regulatory frameworks. The study aims to shed light on investor behavior and the complexities of market reactions in response to misinformation and varying types of news.

Language:

Export as PDF

Export as PDF
Here are 5 key points from the video:

1. **Market Reactions to False Information**: The video discusses how the market can be significantly affected by false information, as demonstrated by historical events such as the false bankruptcy report of United Airlines in 2008 and the misleading announcement regarding Eli Lilly's insulin pricing in 2022. Both incidents illustrate the chaotic impact of misinformation, which can result from social media and old news being resurfaced, leading to drastic market reactions.

2. **Information Sources and Investment Behavior**: The paper explores how information from various sources (companies, analysts, and social media) influences investor beliefs and behavior. It emphasizes the challenge of discerning the accuracy of this information and how it impacts stock prices, suggesting a complex interplay between external information and internal corporate disclosures.

3. **Modeling Market Reactions**: The authors created a mathematical model to analyze how firms decide whether to disclose private information based on external news. They focus on the implications of good vs. bad news and how the timing of these disclosures impacts market responses, revealing that good news can sometimes lead to negative market reactions.

4. **Non-monotonic Price Behavior**: One of the key findings is that market prices can react in non-intuitive ways, particularly in scenarios involving intermediate news. The researchers note that a price drop can occur following favorable news, a reaction stemming from investors' perceptions of the likelihood that managers are informed and the informativeness of the news.

5. **Implications for Empirical Study and Regulation**: The video highlights the need for empirical research to better understand the effects of uncertain informativeness of external news on market behavior. The authors suggest that their findings may raise important questions for regulators and researchers investigating the relationship between external information, corporate disclosure practices, and market dynamics.

Export as PDF
Here are some key quotes extracted from the video, capturing the central themes and important points discussed:

1. **Observation of Market Reactions**:
- "The way we started this thinking about this paper is we observed several very interesting, situations out there in the market."

2. **Impact of Misleading Information**:
- "Now, this caused the market... The company came out relatively quickly, but the market was still unsure."

3. **Role of Social Media**:
- "With the new social media, it just exacerbates everything."

4. **Disinformation Examples**:
- "It turned out, of course, this is completely not true... this information didn't come from Eli Lilly, but it did cause havoc again on the market."

5. **Abundance of Information**:
- "There is abundance of information out there. and how precise or accurate this information is about the firm."

6. **Investor Behavior and Beliefs**:
- "We are very interested in studying this subtle update of beliefs as outside news arise and what happens to investor beliefs, investor behavior, prices that we observe."

7. **Disclosure Strategies**:
- "We try to figure out how the market price is going to look in this situation, how the updating looks like."

8. **Market Perception of News**:
- "Essentially, news that are relatively more favorable are viewed by the market participants as less likely informative or being less informative as opposed to bad news."

9. **Non-Monotonic Price Reactions**:
- "What we do find that in certain regions that's not necessarily the case. And paradoxically, when better news come to market, market participants might react actually negatively."

10. **Timing of Information Disclosure**:
- "More positive news are going to suppress the probability of disclosure. Whereas less favorable news or more negative news are going to exacerbate the probability that the firm is going to disclose."

11. **Effect of Firm's Silence**:
- "When the firm remains silent, what happens is market participants are not... The firm really knew that... but they don't want to disclose it because it's bad."

12. **Credibility of External News**:
- "Investors believe that it is more likely that that manager indeed was informed, but it was also more likely that these external news is indeed informative."

13. **Implications for Regulation**:
- "We believe that our paper or our results are one step forward to explaining why do we observe that... for the regulator to understand what is going on."

These quotes reflect the primary themes of misinformation, market reactions, investor beliefs, and the implications of external news on firm disclosures, all of which are central to the discussed paper.

X-TIMESTAMP-MAP=LOCAL:00:00:00.000,MPEGTS:90000

1
00:00:00.065 --> 00:00:03.045
The way we started this thinking about this paper is we

2
00:00:03.245 --> 00:00:05.445
observed several very interesting,

3
00:00:05.615 --> 00:00:07.685
situations out, out there in the market.

4
00:00:07.865 --> 00:00:09.765
So the first one that we kind

5
00:00:09.765 --> 00:00:13.845
of came across is an event from 2008 when Bloomberg News

6
00:00:14.325 --> 00:00:16.925
reported United Airlines,

7
00:00:17.475 --> 00:00:20.485
that United Airlines are essentially filing for bankruptcy.

8
00:00:20.785 --> 00:00:23.685
Now, this caused the haken the market,

9
00:00:23.745 --> 00:00:24.765
the market was not true.

10
00:00:24.765 --> 00:00:27.325
Whether this is correct or this is not correct.

11
00:00:28.145 --> 00:00:31.205
Uh, the company came out relatively quickly

12
00:00:31.345 --> 00:00:32.565
and said, this is not true.

13
00:00:32.705 --> 00:00:34.125
But the market was still unsure.

14
00:00:34.145 --> 00:00:36.205
Is it really not true or they're trying to escape,

15
00:00:37.065 --> 00:00:40.445
and we saw the price crashing for a while.

16
00:00:40.745 --> 00:00:44.645
Um, so now what it turns out is that somebody, one

17
00:00:44.645 --> 00:00:47.245
of the analysts came upon an old article

18
00:00:47.855 --> 00:00:50.445
about the bankruptcy and literally reposted it.

19
00:00:50.545 --> 00:00:54.405
And from there, with the new social media, it just

20
00:00:55.395 --> 00:00:56.525
exacerbates everything.

21
00:00:56.865 --> 00:00:57.885
And it came out there

22
00:00:58.505 --> 00:01:01.485
and somehow it became the, the,

23
00:01:01.485 --> 00:01:02.725
the biggest news on the market.

24
00:01:03.555 --> 00:01:05.365
Okay? So it turned out, of course,

25
00:01:05.365 --> 00:01:06.485
this is completely not true.

26
00:01:06.515 --> 00:01:07.845
That was a very old article,

27
00:01:07.965 --> 00:01:09.165
I think believe from two thou there,

28
00:01:09.165 --> 00:01:11.005
2002 bankruptcy process.

29
00:01:11.665 --> 00:01:15.525
Um, so, okay, that, that was the first time we came across.

30
00:01:15.785 --> 00:01:19.405
And then more recently, we already were working on

31
00:01:19.405 --> 00:01:22.885
that article in 2022 Company Eli Lilly.

32
00:01:23.265 --> 00:01:25.405
So they, it's a pharmaceutical company.

33
00:01:25.665 --> 00:01:28.365
And then all of a sudden there was this information

34
00:01:28.365 --> 00:01:32.325
that it's going to offer, its, insulin for free.

35
00:01:33.905 --> 00:01:36.365
Now, you can imagine what happened on the market.

36
00:01:36.535 --> 00:01:37.925
Again, all this, it, the,

37
00:01:37.985 --> 00:01:40.205
the information didn't come from Eli Lilly,

38
00:01:40.205 --> 00:01:43.485
but it did cause a havoc again on the market.

39
00:01:44.105 --> 00:01:46.085
Uh, it did turn around

40
00:01:46.085 --> 00:01:48.445
that it was completely untrue in this case.

41
00:01:48.705 --> 00:01:52.285
Uh, there was not even a, a, a grain of, of truth in it.

42
00:01:52.285 --> 00:01:53.325
Whereas with Bloomberg

43
00:01:53.325 --> 00:01:56.085
and Uni United Airlines, it was somehow true,

44
00:01:56.185 --> 00:01:57.725
but it wasn't for that period.

45
00:01:58.785 --> 00:02:01.205
And then we looked for more information.

46
00:02:01.465 --> 00:02:06.085
And then at around 2015, I believe there were two

47
00:02:06.635 --> 00:02:10.285
fake Twitter accounts on behalf of two companies.

48
00:02:11.265 --> 00:02:12.605
One was audience.

49
00:02:12.705 --> 00:02:14.805
I'm trying to look at my notes for the precise names

50
00:02:14.805 --> 00:02:16.805
of the companies, and the other one was

51
00:02:16.835 --> 00:02:17.965
Sarepta Therapeutics.

52
00:02:18.225 --> 00:02:21.445
Now, what what was happening is these were fake accounts,

53
00:02:21.445 --> 00:02:23.565
but of course, from the point of view of the readers

54
00:02:23.565 --> 00:02:27.365
and the social media users, they were completely correct,

55
00:02:28.185 --> 00:02:30.685
and that account was posting information.

56
00:02:30.705 --> 00:02:34.045
It turned out it was a trader who was trying to gain more,

57
00:02:34.145 --> 00:02:37.645
you know, some kind of a profits from out of this.

58
00:02:38.265 --> 00:02:40.565
And here's where we came, we said, okay,

59
00:02:40.715 --> 00:02:42.325
this is very, very interesting.

60
00:02:42.715 --> 00:02:46.245
When there are, like, we think about information that comes

61
00:02:46.305 --> 00:02:47.685
to capital markets,

62
00:02:48.225 --> 00:02:50.045
and we usually, the first thing we think,

63
00:02:50.065 --> 00:02:52.765
at least in my area, is, okay, what did the company

64
00:02:53.285 --> 00:02:54.725
disclose about, or what kind

65
00:02:54.725 --> 00:02:56.205
of information the company communicated,

66
00:02:56.275 --> 00:02:58.445
whether it's in its financial statements

67
00:02:58.865 --> 00:03:01.045
Or whether it's some kind of an additional information

68
00:03:01.045 --> 00:03:03.685
that it chooses voluntary to, to disseminate.

69
00:03:05.025 --> 00:03:06.765
Uh, but in addition

70
00:03:06.765 --> 00:03:09.125
to this information coming from the company,

71
00:03:09.135 --> 00:03:12.685
there is a plet of outside news that are coming,

72
00:03:12.685 --> 00:03:14.565
whether this is analyst forecast,

73
00:03:14.565 --> 00:03:16.925
whether this is coming from media,

74
00:03:17.745 --> 00:03:21.005
and again, media standard, mainstream media

75
00:03:21.345 --> 00:03:24.165
or social media, which is going completely out

76
00:03:24.165 --> 00:03:25.645
of hand, recently.

77
00:03:26.265 --> 00:03:28.005
Um, and, and,

78
00:03:28.025 --> 00:03:30.805
and of course it can be journalists coming out,

79
00:03:31.095 --> 00:03:34.005
peers reporting their own information that we don't know

80
00:03:34.025 --> 00:03:36.325
how correlated it is with the firm

81
00:03:36.325 --> 00:03:37.445
that we are interested in.

82
00:03:37.865 --> 00:03:40.245
So there is abundance of information out there.

83
00:03:40.985 --> 00:03:45.765
And for regular investors, it is unclear how informative

84
00:03:45.825 --> 00:03:48.245
and how precise or accurate this information

85
00:03:48.305 --> 00:03:49.725
is about the firm.

86
00:03:50.265 --> 00:03:54.125
We are very interested in, in studying this subtle update

87
00:03:54.125 --> 00:03:56.885
of beliefs as outside news arise

88
00:03:57.385 --> 00:04:01.565
and what happens to investor beliefs, investor behavior,

89
00:04:02.025 --> 00:04:03.445
prices that we observe.

90
00:04:03.625 --> 00:04:06.125
And that, of course, would have a feedback effect on

91
00:04:06.125 --> 00:04:07.205
what the firm is gonna do.

92
00:04:07.905 --> 00:04:10.565
If the firm is expecting external news

93
00:04:10.565 --> 00:04:12.805
that no one knows whether this is going to be precise,

94
00:04:13.225 --> 00:04:15.565
how is it going to preempt something

95
00:04:15.905 --> 00:04:19.445
and say something more voluntarily If news already come out

96
00:04:19.505 --> 00:04:22.085
and we know, like in United Airlines

97
00:04:22.105 --> 00:04:23.725
and Eddie Lilly news came out,

98
00:04:24.105 --> 00:04:25.525
it probably didn't expect somebody

99
00:04:25.525 --> 00:04:26.805
to come with something so bizarre.

100
00:04:27.185 --> 00:04:28.365
But what do we do now?

101
00:04:28.545 --> 00:04:31.485
Do we come out and say, this isn't true,

102
00:04:32.145 --> 00:04:34.845
or we just shut up and, and wait for the market?

103
00:04:35.025 --> 00:04:38.685
Of course, our conjecture was that, well,

104
00:04:38.685 --> 00:04:40.045
it depends on the positivity

105
00:04:40.265 --> 00:04:42.765
and what is the beliefs about the precision

106
00:04:42.765 --> 00:04:43.925
of the outside information.

107
00:04:43.925 --> 00:04:46.125
Essentially, we are building economic models

108
00:04:46.745 --> 00:04:48.325
and we solve them mathematically.

109
00:04:48.385 --> 00:04:51.805
So the way we did it, we essentially created a somehow

110
00:04:51.945 --> 00:04:54.125
of a simplified version of the world.

111
00:04:54.785 --> 00:04:56.365
Uh, and we solved it.

112
00:04:56.425 --> 00:04:59.485
We made several, we presented several research which,

113
00:04:59.485 --> 00:05:00.965
results, which are propositions,

114
00:05:01.305 --> 00:05:03.285
and then we made predictions,

115
00:05:03.985 --> 00:05:06.605
but into what we expect in the world to happen.

116
00:05:06.745 --> 00:05:10.285
But then afterwards, we kind of try to translate those into

117
00:05:10.635 --> 00:05:13.405
what are we gonna see, what we expect to see in the data.

118
00:05:13.985 --> 00:05:16.325
So if there is an empirical researchers,

119
00:05:16.495 --> 00:05:19.085
those are researchers that look into the real data

120
00:05:19.465 --> 00:05:23.125
and run regulations to figure out association,

121
00:05:23.315 --> 00:05:26.365
sometimes depending on the research design that they said,

122
00:05:26.365 --> 00:05:31.085
they are also able to do, to set kind of causality,

123
00:05:31.355 --> 00:05:33.285
what X causes y for example.

124
00:05:33.425 --> 00:05:36.085
We try to kind of make their work easier

125
00:05:36.385 --> 00:05:40.205
by translating our mathematical results into some kind

126
00:05:40.205 --> 00:05:41.485
of a set of predictions.

127
00:05:41.505 --> 00:05:45.725
In our model, there is a firm and it's, possesses,

128
00:05:45.805 --> 00:05:47.405
but the, there is some probability

129
00:05:47.405 --> 00:05:49.045
that it possesses private information.

130
00:05:49.385 --> 00:05:51.325
It may not possess this private information,

131
00:05:51.825 --> 00:05:55.325
and it essentially can decide whether to disclose it or not.

132
00:05:55.385 --> 00:05:58.125
So this is type of information that's not amenable

133
00:05:58.125 --> 00:06:00.245
for inclusion in financial statements.

134
00:06:00.265 --> 00:06:01.725
For example, financial statements.

135
00:06:01.725 --> 00:06:03.925
We have to disclose this kind of information

136
00:06:03.925 --> 00:06:05.925
and to review it to the market participants.

137
00:06:06.985 --> 00:06:10.205
So there is also an outside source of information,

138
00:06:10.505 --> 00:06:13.245
but a, the source of information may be inaccurate.

139
00:06:13.535 --> 00:06:16.085
There is some probability that it's accurate

140
00:06:16.105 --> 00:06:17.605
and some provided it's inaccurate.

141
00:06:18.185 --> 00:06:20.965
Um, it doesn't need to be perfectly informative

142
00:06:21.105 --> 00:06:22.445
or completely uninformative.

143
00:06:22.575 --> 00:06:25.045
There are some, there is some leeway there.

144
00:06:25.105 --> 00:06:27.965
We, we allow, and it was a little bit of a long process.

145
00:06:28.265 --> 00:06:30.445
Um, it wasn't easy mathematically to do that,

146
00:06:30.505 --> 00:06:33.125
but we are very proud now that we are, have arrived there.

147
00:06:34.305 --> 00:06:38.005
Uh, and then that source of information, the timing where it

148
00:06:38.665 --> 00:06:40.325
it is being revealed is fixed.

149
00:06:40.395 --> 00:06:41.885
When I say source of information,

150
00:06:41.985 --> 00:06:43.445
it really can be an analyst,

151
00:06:43.545 --> 00:06:46.005
it really can be a media source, it can be a,

152
00:06:46.525 --> 00:06:47.725
a social media user.

153
00:06:48.025 --> 00:06:52.285
Um, it can be another firm disclosing its information.

154
00:06:53.025 --> 00:06:56.125
It may be, it may be like,

155
00:06:56.515 --> 00:06:58.005
financial statements of another firm.

156
00:06:58.065 --> 00:07:01.125
So that firm is, has to disclose it, but it may

157
00:07:01.125 --> 00:07:02.365
or may not be related to us

158
00:07:02.425 --> 00:07:05.565
or with certain correl somehow correlated to the value.

159
00:07:06.585 --> 00:07:11.365
So, the main model looks at whether,

160
00:07:11.515 --> 00:07:15.845
essentially whether the firm can, the firm is disclosing

161
00:07:15.845 --> 00:07:17.525
before in an anticipation

162
00:07:17.525 --> 00:07:19.405
of this external news to be coming.

163
00:07:19.855 --> 00:07:22.045
Let's call it the peers financial statement.

164
00:07:22.945 --> 00:07:24.605
So when can it happen?

165
00:07:24.715 --> 00:07:29.285
This can happen, for example, when a firm, is,

166
00:07:29.385 --> 00:07:31.045
has a scheduled conference call

167
00:07:31.225 --> 00:07:33.445
before the anticip based anticipated release

168
00:07:33.465 --> 00:07:35.205
of this peers financial statement.

169
00:07:35.945 --> 00:07:39.565
Or it can also happen when there is an anticipated,

170
00:07:39.575 --> 00:07:42.885
quiet period in the lead up to an initial public offering

171
00:07:43.185 --> 00:07:46.965
or, the, the business end, results.

172
00:07:47.185 --> 00:07:48.845
And then we extend and we say,

173
00:07:48.845 --> 00:07:51.405
what will happen if it can be later on?

174
00:07:51.505 --> 00:07:53.245
We relax. We, we relax that

175
00:07:53.245 --> 00:07:55.765
and say, what happens if the news just come up

176
00:07:55.865 --> 00:07:58.365
and now we need to decide whether to respond or not?

177
00:07:59.425 --> 00:08:00.885
Um, and

178
00:08:00.885 --> 00:08:02.645
that's precisely the setting that we're trying to solve.

179
00:08:02.745 --> 00:08:05.405
We are trying to figure out how the market price is going

180
00:08:05.465 --> 00:08:08.725
to look in this situation, how, how the updating looks like.

181
00:08:09.265 --> 00:08:13.765
And then because the firm is trying to maximize its mi

182
00:08:13.765 --> 00:08:16.525
market price, we are going to see how it's going,

183
00:08:16.625 --> 00:08:19.245
how it is going to react in anticipation

184
00:08:19.545 --> 00:08:22.885
or in response to such news, later on.

185
00:08:22.955 --> 00:08:24.365
Essentially, news

186
00:08:24.435 --> 00:08:28.085
that are relatively more favorable are viewed

187
00:08:28.185 --> 00:08:31.685
by the market participants as less likely informative

188
00:08:31.785 --> 00:08:35.165
or being less informative as opposed to bad news.

189
00:08:35.185 --> 00:08:37.965
So somehow the market perceives bad news

190
00:08:38.465 --> 00:08:39.565
as being more accurate.

191
00:08:40.785 --> 00:08:45.245
Um, and this is the first finding,

192
00:08:45.425 --> 00:08:48.085
let me briefly summarize the, the continuation

193
00:08:48.085 --> 00:08:49.605
and then I'll go back and explain the

194
00:08:49.605 --> 00:08:50.885
mechanism through which it works.

195
00:08:51.555 --> 00:08:55.445
What this causes is that there is some kind of a,

196
00:08:55.465 --> 00:08:57.605
the market price is going to be non monotonic.

197
00:08:57.985 --> 00:09:01.005
We always think that better news should be bringing up,

198
00:09:01.105 --> 00:09:02.125
a higher price.

199
00:09:02.425 --> 00:09:04.805
But what we do find that in certain regions

200
00:09:04.805 --> 00:09:06.165
that's not necessarily the case.

201
00:09:06.505 --> 00:09:09.005
And paradoxically, when better news come to market,

202
00:09:10.025 --> 00:09:12.765
market participants might react actually negatively.

203
00:09:12.765 --> 00:09:15.805
And we are gonna see this, downward reaction.

204
00:09:16.705 --> 00:09:19.805
Uh, furthermore, we try to figure out what would happen.

205
00:09:20.145 --> 00:09:24.965
Uh, the more we anticipate that this news, external news is,

206
00:09:25.305 --> 00:09:27.645
more informative,

207
00:09:28.105 --> 00:09:31.645
and we find that, well overall,

208
00:09:31.645 --> 00:09:34.205
it really depends on the timing of the disclosures.

209
00:09:34.205 --> 00:09:38.445
We are disclosing before that, we, we are going

210
00:09:38.445 --> 00:09:41.045
to see an overall suppression in the probability

211
00:09:41.045 --> 00:09:44.205
that the firm is going to disclose if this disclosure is

212
00:09:44.205 --> 00:09:46.605
happening after the release of external news.

213
00:09:47.075 --> 00:09:51.245
Whether exon anticipation of the news

214
00:09:51.265 --> 00:09:54.565
to be more informative is going to increase

215
00:09:54.705 --> 00:09:57.725
or decrease the probability of disclosure would depend on

216
00:09:57.745 --> 00:09:59.365
how positive these news are.

217
00:09:59.875 --> 00:10:03.285
More positive news are going to suppress,

218
00:10:03.305 --> 00:10:04.605
the probability of disclosure.

219
00:10:04.995 --> 00:10:07.485
Whereas less favorable news

220
00:10:07.505 --> 00:10:11.125
or more negative news are going to, kind

221
00:10:11.125 --> 00:10:13.885
of e exacerbate the probability

222
00:10:13.885 --> 00:10:15.405
that the firm is going to disclose.

223
00:10:15.555 --> 00:10:18.925
What is the equilibrium in general when those type of games,

224
00:10:19.075 --> 00:10:21.445
when firms can disclose some kind of information

225
00:10:21.745 --> 00:10:23.005
or they can withhold it.

226
00:10:23.705 --> 00:10:27.525
If we know, if market participants know which certainty

227
00:10:27.525 --> 00:10:29.005
that the firm is informed,

228
00:10:31.195 --> 00:10:34.605
seminal papers coming, going back to the eighties, are going

229
00:10:34.605 --> 00:10:36.965
to tell us, well, this information is going to unravel.

230
00:10:36.965 --> 00:10:39.005
If there is no friction out there on the market,

231
00:10:39.675 --> 00:10:41.085
information will unravel.

232
00:10:41.435 --> 00:10:45.405
Well, it's not the case when market participants are unsure

233
00:10:45.405 --> 00:10:47.685
whether the firm has information or not,

234
00:10:47.915 --> 00:10:50.085
because when the firm remains silent,

235
00:10:50.795 --> 00:10:52.845
what happens is market participants are not,

236
00:10:52.845 --> 00:10:55.405
are not sure whether it was because these were bad news.

237
00:10:55.425 --> 00:10:57.845
The firm really knew that, knew the news,

238
00:10:57.945 --> 00:11:00.445
but they don't want to disclose it because it's bad,

239
00:11:00.825 --> 00:11:03.325
or it is because the, the firm simply doesn't know anything.

240
00:11:04.545 --> 00:11:07.685
So this is also what is going on in our model,

241
00:11:08.025 --> 00:11:11.125
but now we also combine it with the fact that there is this

242
00:11:11.805 --> 00:11:14.365
external news that we don't know whether it's

243
00:11:14.505 --> 00:11:16.245
how informative, precise it is.

244
00:11:16.745 --> 00:11:21.325
So, yeah, on average we know

245
00:11:21.325 --> 00:11:24.245
that the firm is going to disclose and review better news

246
00:11:24.505 --> 00:11:26.005
and withhold towards news,

247
00:11:26.825 --> 00:11:29.805
but it also kind of, is going

248
00:11:29.825 --> 00:11:34.285
to endogenously in be intertwined with how investors believe

249
00:11:34.755 --> 00:11:38.165
what we think about, the external news informativeness,

250
00:11:38.275 --> 00:11:40.205
even though exactly we assume then

251
00:11:40.205 --> 00:11:41.485
to be completely uncorrelated,

252
00:11:41.515 --> 00:11:46.085
they endogenously become really co intertwined now

253
00:11:46.995 --> 00:11:51.205
when external participants, investors observe some piece

254
00:11:51.205 --> 00:11:54.285
of information that is negative,

255
00:11:54.835 --> 00:11:57.445
some more negative information about the firm.

256
00:11:57.985 --> 00:11:59.965
And at the same time, the manager

257
00:12:00.505 --> 00:12:02.005
of the firm remains silent.

258
00:12:03.035 --> 00:12:06.725
Investors believe that it is more likely that

259
00:12:06.725 --> 00:12:08.885
that manager indeed was informed,

260
00:12:10.755 --> 00:12:13.375
but it was also more likely that these external news

261
00:12:14.115 --> 00:12:15.735
is indeed informative.

262
00:12:17.165 --> 00:12:18.985
Now, on contrary when,

263
00:12:19.285 --> 00:12:21.265
or more informative on the contrary,

264
00:12:21.265 --> 00:12:23.705
when the market observes a more positive,

265
00:12:24.015 --> 00:12:25.985
more positive news,

266
00:12:26.125 --> 00:12:30.225
and the manager still remains silent, the market views it

267
00:12:30.225 --> 00:12:32.545
as more likely that that manager knew nothing.

268
00:12:32.545 --> 00:12:35.865
Because if that manager was informed were to be informed,

269
00:12:36.375 --> 00:12:40.585
that manager would have disclosed this new, this news,

270
00:12:40.735 --> 00:12:42.945
himself or herself or, or, or whatever.

271
00:12:44.955 --> 00:12:47.695
Uh, so this is, this is the main driving force in the

272
00:12:48.295 --> 00:12:49.335
updating of the beliefs.

273
00:12:49.835 --> 00:12:52.895
So once there is a certain specific value

274
00:12:52.895 --> 00:12:55.015
where the marketing equilibrium expects the manager

275
00:12:55.075 --> 00:12:58.535
to start talking, if that manager were was informed,

276
00:12:58.915 --> 00:13:01.855
so a rounded value, if the news

277
00:13:01.995 --> 00:13:04.215
as the news become more positive, all

278
00:13:04.215 --> 00:13:06.655
of a sudden there is a drop in the beliefs of the market

279
00:13:06.685 --> 00:13:09.855
that that that about the joint probability

280
00:13:09.855 --> 00:13:14.375
that this news is informative and the manager is informed.

281
00:13:14.525 --> 00:13:16.535
This is really the main driving force

282
00:13:16.635 --> 00:13:18.615
behind the non monotonicity of the price.

283
00:13:18.615 --> 00:13:21.615
Because for when the news is really, really negative

284
00:13:21.675 --> 00:13:24.655
or very, very positive, the prices will behave exactly

285
00:13:24.755 --> 00:13:26.655
as we expect them to behave.

286
00:13:27.275 --> 00:13:29.855
But as we go to this intermediate region

287
00:13:30.085 --> 00:13:32.295
that these drop in beliefs is going

288
00:13:32.355 --> 00:13:37.095
to cause this sharp drop in the price, we have observed in

289
00:13:38.455 --> 00:13:41.505
empirical studies, a lot of confounding information.

290
00:13:41.725 --> 00:13:42.745
So, in results.

291
00:13:42.765 --> 00:13:45.065
So essentially we would see whether price

292
00:13:45.385 --> 00:13:46.545
increase or price decrease.

293
00:13:46.565 --> 00:13:49.705
So there is kind of a conflicting ev empirical evidence

294
00:13:49.705 --> 00:13:51.185
that we are, that we are observing.

295
00:13:51.605 --> 00:13:54.345
So we believe that our paper

296
00:13:54.605 --> 00:13:57.225
or our results are one step forward

297
00:13:57.365 --> 00:13:59.425
to explaining why do we observe that?

298
00:13:59.805 --> 00:14:03.265
So we need to be, per, what we suggest is that empirical

299
00:14:03.785 --> 00:14:05.545
researchers need to be accounting

300
00:14:05.565 --> 00:14:09.425
for this uncertain informativeness of external news

301
00:14:09.965 --> 00:14:14.025
and be, it might not be extremely easy to, to do that

302
00:14:14.025 --> 00:14:15.945
because how do I set

303
00:14:15.965 --> 00:14:19.025
and how do I control it, where precisely is the part

304
00:14:19.025 --> 00:14:20.345
where I expect this drop,

305
00:14:20.685 --> 00:14:24.505
but our work still will, is one step forward in terms

306
00:14:24.645 --> 00:14:27.945
of explaining this and raising the flag

307
00:14:27.965 --> 00:14:30.265
and saying, okay, here is something maybe we are missing.

308
00:14:30.365 --> 00:14:32.785
Can we to be studying one step forward also

309
00:14:32.785 --> 00:14:35.185
for the regulator to understand what is going on

310
00:14:35.205 --> 00:14:37.745
and when market prices are behaving unexpectedly,

311
00:14:37.965 --> 00:14:40.545
to be aware that this might be the reason

312
00:14:40.685 --> 00:14:45.465
behind it in the wake of all kinds of news that we are,

313
00:14:45.525 --> 00:14:46.065
around us.

Market Misinformation: Case Studies of United Airlines and Eli Lilly

00:00:00 - The way we started this thinking about this paper is we observed several very interesting, situations out, out there in the market. So the first one that we kind of came across is an event from 2008 when Bloomberg News reported United Airlines, that United Airlines are essentially filing for bankruptcy. Now, this caused the haken the market, the market was not true. Whether this is correct or this is not correct. the company came out relatively quickly

00:00:32 - But the market was still unsure. Is it really not true or they're trying to escape, and we saw the price crashing for a while. so now what it turns out is that somebody, one of the analysts came upon an old article about the bankruptcy and literally reposted it. And from there, with the new social media, it just exacerbates everything. And it came out there and somehow it became the, the,

00:01:03 - Okay? So it turned out, of course, this is completely not true. That was a very old article, I think believe from two thou there, 2002 bankruptcy process. so, okay, that, that was the first time we came across. And then more recently, we already were working on that article in 2022 Company Eli Lilly. So they, it's a pharmaceutical company. And then all of a sudden there was this information that it's going to offer, its, insulin for free.

Market Chaos from Misinformation: Case Studies on Eli Lilly and United Airlines

00:01:36 - Again, all this, it, the, the information didn't come from Eli Lilly, but it did cause a havoc again on the market. it did turn around that it was completely untrue in this case. there was not even a, a, a grain of, of truth in it. Whereas with Bloomberg and Uni United Airlines, it was somehow true, but it wasn't for that period. And then we looked for more information. And then at around 2015, I believe there were two

00:02:11 - One was audience. I'm trying to look at my notes for the precise names of the companies, and the other one was Sarepta Therapeutics. Now, what what was happening is these were fake accounts, but of course, from the point of view of the readers and the social media users, they were completely correct, and that account was posting information. It turned out it was a trader who was trying to gain more, you know, some kind of a profits from out of this.

00:02:40 - this is very, very interesting. When there are, like, we think about information that comes to capital markets, and we usually, the first thing we think, at least in my area, is, okay, what did the company disclose about, or what kind of information the company communicated, whether it's in its financial statements Or whether it's some kind of an additional information that it chooses voluntary to, to disseminate. but in addition to this information coming from the company,

Challenges of Information Accuracy in Market Responses

00:03:12 - whether this is analyst forecast, whether this is coming from media, and again, media standard, mainstream media or social media, which is going completely out of hand, recently. and, and, and of course it can be journalists coming out, peers reporting their own information that we don't know how correlated it is with the firm that we are interested in. So there is abundance of information out there.

00:03:45 - and how precise or accurate this information is about the firm. We are very interested in, in studying this subtle update of beliefs as outside news arise and what happens to investor beliefs, investor behavior, prices that we observe. And that, of course, would have a feedback effect on what the firm is gonna do. If the firm is expecting external news that no one knows whether this is going to be precise,

00:04:15 - and say something more voluntarily If news already come out and we know, like in United Airlines and Eddie Lilly news came out, it probably didn't expect somebody to come with something so bizarre. But what do we do now? Do we come out and say, this isn't true, or we just shut up and, and wait for the market? Of course, our conjecture was that, well, it depends on the positivity and what is the beliefs about the precision of the outside information.

Mathematical Modeling of Information Disclosure in Firms

00:04:46 - and we solve them mathematically. So the way we did it, we essentially created a somehow of a simplified version of the world. and we solved it. We made several, we presented several research which, results, which are propositions, and then we made predictions, but into what we expect in the world to happen. But then afterwards, we kind of try to translate those into what are we gonna see, what we expect to see in the data.

00:05:16 - those are researchers that look into the real data and run regulations to figure out association, sometimes depending on the research design that they said, they are also able to do, to set kind of causality, what X causes y for example. We try to kind of make their work easier by translating our mathematical results into some kind of a set of predictions. In our model, there is a firm and it's, possesses,

00:05:47 - that it possesses private information. It may not possess this private information, and it essentially can decide whether to disclose it or not. So this is type of information that's not amenable for inclusion in financial statements. For example, financial statements. We have to disclose this kind of information and to review it to the market participants. So there is also an outside source of information, but a, the source of information may be inaccurate. There is some probability that it's accurate

Understanding Information Disclosure in Market Contexts

00:06:18 - it doesn't need to be perfectly informative or completely uninformative. There are some, there is some leeway there. We, we allow, and it was a little bit of a long process. it wasn't easy mathematically to do that, but we are very proud now that we are, have arrived there. and then that source of information, the timing where it it is being revealed is fixed. When I say source of information, it really can be an analyst, it really can be a media source, it can be a,

00:06:48 - it can be another firm disclosing its information. It may be, it may be like, financial statements of another firm. So that firm is, has to disclose it, but it may or may not be related to us or with certain correl somehow correlated to the value. So, the main model looks at whether, essentially whether the firm can, the firm is disclosing before in an anticipation

00:07:19 - Let's call it the peers financial statement. So when can it happen? This can happen, for example, when a firm, is, has a scheduled conference call before the anticip based anticipated release of this peers financial statement. Or it can also happen when there is an anticipated, quiet period in the lead up to an initial public offering or, the, the business end, results. And then we extend and we say,

Market Reaction Dynamics: The Paradox of Positive News

00:07:51 - We relax. We, we relax that and say, what happens if the news just come up and now we need to decide whether to respond or not? and that's precisely the setting that we're trying to solve. We are trying to figure out how the market price is going to look in this situation, how, how the updating looks like. And then because the firm is trying to maximize its mi market price, we are going to see how it's going, how it is going to react in anticipation

00:08:22 - Essentially, news that are relatively more favorable are viewed by the market participants as less likely informative or being less informative as opposed to bad news. So somehow the market perceives bad news as being more accurate. and this is the first finding, let me briefly summarize the, the continuation and then I'll go back and explain the

00:08:51 - What this causes is that there is some kind of a, the market price is going to be non monotonic. We always think that better news should be bringing up, a higher price. But what we do find that in certain regions that's not necessarily the case. And paradoxically, when better news come to market, market participants might react actually negatively. And we are gonna see this, downward reaction. furthermore, we try to figure out what would happen.

Timing and Impact of Information Disclosure in Market Dynamics

00:09:25 - more informative, and we find that, well overall, it really depends on the timing of the disclosures. We are disclosing before that, we, we are going to see an overall suppression in the probability that the firm is going to disclose if this disclosure is happening after the release of external news. Whether exon anticipation of the news

00:09:54 - or decrease the probability of disclosure would depend on how positive these news are. More positive news are going to suppress, the probability of disclosure. Whereas less favorable news or more negative news are going to, kind of e exacerbate the probability that the firm is going to disclose. What is the equilibrium in general when those type of games, when firms can disclose some kind of information

00:10:23 - If we know, if market participants know which certainty that the firm is informed, seminal papers coming, going back to the eighties, are going to tell us, well, this information is going to unravel. If there is no friction out there on the market, information will unravel. Well, it's not the case when market participants are unsure whether the firm has information or not, because when the firm remains silent, what happens is market participants are not,

Firm Disclosure Dynamics and Investor Perceptions

00:10:55 - The firm really knew that, knew the news, but they don't want to disclose it because it's bad, or it is because the, the firm simply doesn't know anything. So this is also what is going on in our model, but now we also combine it with the fact that there is this external news that we don't know whether it's how informative, precise it is. So, yeah, on average we know that the firm is going to disclose and review better news

00:11:26 - but it also kind of, is going to endogenously in be intertwined with how investors believe what we think about, the external news informativeness, even though exactly we assume then to be completely uncorrelated, they endogenously become really co intertwined now when external participants, investors observe some piece of information that is negative,

00:11:57 - And at the same time, the manager of the firm remains silent. Investors believe that it is more likely that that manager indeed was informed, but it was also more likely that these external news is indeed informative. Now, on contrary when, or more informative on the contrary, when the market observes a more positive, more positive news,

Investor Belief Dynamics and Market Response to Information

00:12:30 - as more likely that that manager knew nothing. Because if that manager was informed were to be informed, that manager would have disclosed this new, this news, himself or herself or, or, or whatever. so this is, this is the main driving force in the updating of the beliefs. So once there is a certain specific value where the marketing equilibrium expects the manager to start talking, if that manager were was informed,

00:13:01 - as the news become more positive, all of a sudden there is a drop in the beliefs of the market that that that about the joint probability that this news is informative and the manager is informed. This is really the main driving force behind the non monotonicity of the price. Because for when the news is really, really negative or very, very positive, the prices will behave exactly as we expect them to behave. But as we go to this intermediate region

00:13:32 - to cause this sharp drop in the price, we have observed in empirical studies, a lot of confounding information. So, in results. So essentially we would see whether price increase or price decrease. So there is kind of a conflicting ev empirical evidence that we are, that we are observing. So we believe that our paper or our results are one step forward to explaining why do we observe that? So we need to be, per, what we suggest is that empirical

Navigating the Challenges of External News Informativeness

00:14:05 - for this uncertain informativeness of external news and be, it might not be extremely easy to, to do that because how do I set and how do I control it, where precisely is the part where I expect this drop, but our work still will, is one step forward in terms of explaining this and raising the flag and saying, okay, here is something maybe we are missing. Can we to be studying one step forward also for the regulator to understand what is going on

00:14:45 - to be aware that this might be the reason behind it in the wake of all kinds of news that we are, around us.

Faculti AI can make mistakes

Research

Policy

Analysis

Data

Originals

Explainers

Podcasts

Faculti

New on Faculti

Idea Explorer

Definition

News Context

Critical Questions and Answers

Research Summaries

With a Grain of Salt: Uncertain News and Firms Disclosures

Media Literacy Toolkit

Critical Questions to Consider ▼

Bias and Perspective Awareness ▼

Further Reading ▼