sustainability

Article
Private Information Dissemination and Noise Trading:
Implications for Price Efficiency and Market Liquidity
Huan Liu 1 , Weiqi Liu 1, * and Yi Li 2

1 School of Economic and Management, Shanxi University, Taiyuan 030006, China

2 School of Statistics, Shanxi University of Finance and Economics, Taiyuan 030006, China
* Correspondence: [email protected]; Tel.: +86-137-0358-5581

Abstract: Information is the basis for the sustainable and stable development of financial markets.
Advanced internet technology has accelerated the dissemination of information. To investigate
the impacts of private information dissemination on the sustainability of the financial market, we
construct a rational expectation equilibrium (REE) model. The dissemination of private information
favors noise traders who receive private information and weakens the advantage of informed traders
who have direct access to private information, thus reducing noise-driven volatility and increasing
fundamental-driven volatility, which is not conducive to the sustainability and stability of the financial
market. Private information dissemination increases information asymmetry, reduces the number of
noise traders in the market, decreases market liquidity, and hurts price efficiency for both exogenous
and endogenous information acquisition, which is harmful to the sustainability of the financial market.
Additionally, we numerically analyze the effects of private information on noise traders, market
liquidity, and price efficiency. The numerical results are consistent with the theoretical analysis. The
findings highlight the potential of private information dissemination to noise traders in financial



market analysis. This study contributes to the analysis of financial market sustainability.


Citation: Liu, H.; Liu, W.; Li, Y.
Keywords: information acquisition; information dissemination; price efficiency; market liquidity
Private Information Dissemination
and Noise Trading: Implications for
Price Efficiency and Market Liquidity.
Sustainability 2022, 14, 11624.
https://doi.org/10.3390/ 1. Introduction
su141811624 Information is an important issue in financial market research. Improvements in
Academic Editor: Klaus Reiner
information technology allow information to be disseminated faster and the dissemination
Schenk-Hoppé
range to be wider. In such an environment, information dissemination efficiency is im-
proved. Information disseminates among traders, including noise traders [1], which creates
Received: 26 July 2022 a challenge for the financial market supervision authorities and for the sustainability of
Accepted: 8 September 2022 the financial market. Price volatility, market liquidity, and price efficiency are important
Published: 16 September 2022 factors affecting the sustainability of financial markets ([2–5]). An in-depth study of those
Publisher’s Note: MDPI stays neutral issues could help to formulate effective regulatory policies for the sustainable development
with regard to jurisdictional claims in of financial markets.
published maps and institutional affil- Starting with Grossman and Stiglitz [6], most research in this area is based on a prior
iations. assumption that noise traders are irrational and that all noise traders make the same trading
decisions with information unrelated to fundamental value ([6–8]). However, in the real
economy, noise traders are unwilling to pay a cost to acquire information or falsely think
that they have unique special sources of useful information. Noise traders will not refuse
Copyright: © 2022 by the authors. free private information. Private information will be considered in the trading decision
Licensee MDPI, Basel, Switzerland.
when a noise trader receives private information without cost ([9–12]). Thus, noise trading
This article is an open access article
may be influenced by private information dissemination. In this paper, we aim to improve
distributed under the terms and
our understanding of how noise traders affect financial markets. To this end, we take the
conditions of the Creative Commons
influence of private information dissemination on noise traders into account.
Attribution (CC BY) license (https://
We construct a REE model to explore the implications of private information dissemi-
creativecommons.org/licenses/by/
4.0/).
nation for the financial market. Two assets exist in a competitive market: a risky asset and

Sustainability 2022, 14, 11624. https://doi.org/10.3390/su141811624 https://www.mdpi.com/journal/sustainability

Sustainability 2022, 14, 11624 2 of 19

a risk-free asset. The risky asset is of limited supply, while the risk-free asset is infinitely
supplied. Private information is only available to some traders. Informed traders are
traders who own private information. They make trading based on the public information
and private information that they own. Noise traders make their trading decisions via an
optimal trade-off between the expected loss from trading against informed traders and
the liquidity benefit of market participation. Private information may be disseminated
between close family and friends. When an informed trader owns private information,
they might share it with some noise traders. In this way, these uninformed noise traders
will change state to being informed. All the noise traders are divided into two groups. The
first group of noise traders are senior noise traders. Senior noise traders are noise traders
receiving disseminated private information from informed traders. The second group are
the remaining uninformed noise traders; we call them junior noise traders. Direct interper-
sonal communication is important in traders’ decision-making [13]. As a result of receiving
private information, the demand functions of such noise traders will be different from
those who have not received any private information. With different demand functions,
senior noise traders and junior noise traders have different effects on the market. A trading
fluctuation will be caused by the change of noise traders from junior to senior.
In summary, we find that private information dissemination has implications for price
volatility, price efficiency, and market liquidity. Indeed, private information dissemination
increases information asymmetry, leads fewer noise traders to be active, and lowers market
liquidity, which impedes the sustainable development of the financial market ([3,4]). Private
information is one of the factors behind price volatility [14]. Private information dissemina-
tion aggravates price volatility and harms price efficiency by attracting uninformed trading,
which has a negative impact on the sustainable development of the financial market ([5,15]).
Therefore, the dissemination of private information to noise traders is disadvantageous to
the sustainability of the financial market.
Our findings on market liquidity and price efficiency contrast with those of
Han et al. [8], who report the influence of public information on price efficiency and
market liquidity. The study analyzed the impacts of public information precision on finan-
cial markets under the assumption that there is no private information dissemination. The
normal state of the economy, involving information asymmetry and private information
dissemination, should be used to truly reflect the information and efficiency of the financial
market. Private information can be disseminated from informed traders to some noise
traders for various reasons, such as emotional motivations, social information, etc. In
our research, we take this private information dissemination into account and find that
Han et al. [8] overestimate the influence of public information on noise traders’ number
and market liquidity, as well as underestimating the influence of public information on
price efficiency.
The financial markets relate to sustainable development [16]. The sustainability of
the financial market relates to its ability to retain its identity or its capability to perform
its functions. The sustainable and stable development of the financial market provides a
financial guarantee for the sustainable development of the society and the environment.
A better understanding of the drivers behind the price volatility and market liquidity in
financial markets will contribute to a better analysis of financial market sustainability. In
this sense, our study contributes to the analysis of sustainability.
The remainder of this paper is organized as follows. Section 2 provides related studies.
Section 3 is the model description and assumptions. The equilibrium and implications
of private information dissemination are discussed in Section 4. Section 5 extends the
economy to the case where private information is endogenously acquired. Section 6
presents a simulation to verify the theoretical analysis. Section 7 concludes this work.

2. Literature Review
Our study relates to literature that examines the effect of information dissemination
on the financial market. Information dissemination improves price efficiency when in-
Sustainability 2022, 14, 11624 3 of 19

formation is obtained exogenously ([17,18]) but reduces it when information is acquired

endogenously ([7,19]). The dissemination of information between rational traders improves
market liquidity ([1,20]). Colla and Mele [17] propose an equilibrium model for the impacts
of public and private information on financial markets. Van Ness et al. [21] analyze the
relationship between private information in trades and fluctuations in stock prices and
demonstrate that trading-conveyed private information reduces stock return synchronicity.
Grego [22] finds that the dissemination of information about price volatility decreases price
efficiency. A better understanding of the liquidity risk leads to sustainable trading for
traders [4]. Noise traders, as liquidity providers and price volatility drivers, play an impor-
tant role in the sustainable and stable development of financial markets ([3,4]). However,
existing studies have not examined the effect of private information dissemination on noise
traders, and our study can fill this gap.
There are studies that have analyzed the way information is disseminated in financial
markets. Starting with Admati and Pfleiderer [23], private information is disseminated
from informed traders to uninformed traders through price in a great deal of research. Thus,
price is not only an aggregator of information but also a transmitter ([24,25]). The price acts
as a linker connecting traders’ own information with the information already contained in
the price. Several studies investigate the way in which information disseminates among
traders through more direct channels ([7,9,20,24,26,27], etc.). Bowden and McDonald [28]
research information sharing with social networks and find that expected social learning
relates to the experts. Han and Yang [7] study the implications of social communication
for market efficiency and liquidity. Hong et al. [9] prove that social interaction helps to
increase stock market participation. Ahern [20] uses hand-collected data, finding that inside
information flows through strong ties based on family, friends, and geographic proximity.
In Foerster [27] and Andrei and Cujean [29], private information is disseminated through
social networks through talking. Information dissemination in the above studies only takes
place between rational traders. However, none of these studies consider the channel of
private information dissemination to noise traders.
Private information being widely disseminated among families and friends who
have intimate relationships is a pervasive phenomenon. Thus, it is possible that private
information disseminates from an informed trader to some noise traders who are intimate.
With different expectations, the demand functions of noise traders who receive private
information should be different from those who do not receive private information. We take
the influence of information dissemination on noise trading into consideration and build
a REE model to study its impact on the financial market. Previous studies suggest that
private information dissemination among rational traders increases market liquidity [20]
and price efficiency ([17,18]). Our study on private information dissemination comes to
different conclusions, indicating that the dissemination of private information to noise
traders decreases market liquidity and price efficiency. Our paper adds to the growing
body of literature on information dissemination in financial markets.
Noise makes financial markets possible [30]. An increase in noise trading can improve
market liquidity ([8,23,31,32]) but may reduce price efficiency ([8,31–33]). However, most
studies adopt the definition of noise traders given by Kyle [34] and Black [30]. In these stud-
ies, noise trading is based on information that is unrelated to fundamental value ([35,36]).
Yuan et al. [37] define noise traders as traders who trade based on incomplete information.
Noise traders are not completely irrational traders, but rather bounded rational traders
with a low degree of rationality ([38–40]). A trader profits not because they have better
information but because they have some information that has not been discounted by the
market [41]. A trader’s expected trading profit depends on the information that they own
and information reflected in the price. As private information has not been absorbed by
the price, when a noise trader receives private information for some reason, the private
information will affect the noise trader’s expecting trading profit. However, the influence of
private information dissemination on noise traders has not been considered in the existing
Sustainability 2022, 14, 11624 4 of 19

studies. We shed light on the role that private information plays in the decision-making
process of noise traders.
Grossman and Stiglitz [6] developed a REE model that is widely used to study the
motivations for acquiring costly information and its impact on the financial market. Traders’
trading decisions are made based on a trade-off depending on all available information,
whereas Grossman and Stiglitz [6] ignore information outside the price channel. Following
this, a growing body of literature has used the REE model to analyze the impact of informa-
tion on the financial market ([42–45]). However, none of these researchers considered the
impact of the dissemination of private information to noise traders. We study the impact
of private information transfer on noise traders and further analyze the impact of private
information dissemination on price volatility, market liquidity, and price efficiency. We
revise the model in Han et al. [8] and consider information dissemination in it. Colla and
Mele [17] and Han N. O. [18] prove that social communication improves price efficiency
when information is exogenous. However, our results show that, in both endogenous and
exogenous cases, information dissemination negatively affects price efficiency. Therefore,
our paper complements Colla and Mele [17] and Han N. O. [18].

3. Model Description and Assumptions

Suppose there are two assets in a competitive market: a risky asset and a risk-free
asset. The risky asset has an endogenous price p̃ and a fixed supply. Simply, the risky asset
supply is normalized as one share. The risk-free asset is unlimited and supplied with a
constant value of 1. Traders who decide to participate in the emerging trading market are
called active traders, and those who do not participate in the market are called inactive
traders. The active traders will be divided into three types: informed traders, senior noise
traders, and junior noise traders.
All traders can observe public signal ỹ and asset price p̃, where

ỹ = ṽ + η̃, with η̃ ∼ N (0, 1/ρη ) and ρη > 0. (1)

ρη is the precision of public signal ỹ; a larger ρη means that ỹ contains more information
regarding future payoff.
Suppose that the informed traders are continuous in [0, 1]. All the informed traders
have a constant absolute risk aversion (CARA) utility; the risk aversion coefficient is
denoted by γ > 0. An informed trader i not only can observe the public information from
a public signal ỹ and the price p̃, but they also have access to private information from a
private signal s˜i . s˜i satisfies the following form:

s̃i = ṽ + ε̃ i , with ε̃ i ∼ N (0, 1/ρε i ) and ρε i > 0. (2)

Informed traders trade on public information (ỹ, p̃) and their private information s̃i to
maximize their expected utility.
Noise traders have mastered some information irrelevant to future payoff, but they
believe they have all of the information regarding the future payoff of the risk asset.
Noise traders are unlikely to acquire private information actively, but they will not refuse
free information. Noise traders trade on all of the information they own to maximize
their benefit. Therefore, if a noise trader receives private information disseminated from
informed traders, the trading of the noise trader will be affected. Noise traders are not
perfectly rational traders, and their learning and analysis abilities are weaker than informed
traders. Hence, though they receive disseminated private information, noise traders have a
different demand function from informed traders.
The time is divided into three periods. At t = 0, noise traders can only obtain
public information from ỹ and p̃. Between t = 0 and t = 1, private information can be
disseminated from informed traders to parts of the noise traders for some reasons, such as
emotional motivations and social information, etc. Noise traders make trading decisions
at t = 1. Suppose that the private information dissemination between t = 1 and t = 2
Sustainability 2022, 14, 11624 5 of 19

has no effects on noise traders’ decisions. After the dissemination of private information
during the first half period, a part of the noise traders become informed with a probability
q (0 6 q 6 1). These noise traders are named senior noise traders. They are different
from informed traders in demand functions. Noise traders who have not received private
information until t = 1 are called junior noise traders. At t = 1, noise traders trade-off
between the costs and benefits, and then make decisions regarding whether to participate in
the trading and how much to trade. At t = 2, active traders trade their assets competitively.
Cash flows ṽ are realized at t = 3.
In Admati and Pfleiderer [23], noise trader l who decides to trade must trade a fixed
amount of risk asset. In our analysis, demands are different for junior and senior noise
traders. Junior noise traders trade like the noise traders in Admati and Pfleiderer [23].
However, for the senior traders, the private information leads them to trade δ̃ more units of
a risk asset; δ̃ can be both positive or negative.
By participating in the market, active noise traders can enjoy the exogenous liquidity
benefit B > 0. Assuming that the expected utility of a noise trader is W, then W can be
positive or negative, since there are endogenous trading costs.
The number of active noise traders in the market is defined as L. Let x̃l be the share of
risk assets traded by a noise trader l, l ∈ [0, L]. Then, x̃l has the following form:

ũ + z̃l , junior noise traders,

x̃l = { (3)
ũ + δ̃ + z̃l , senior noise traders,

where ũ ∼ N (0, 1), δ̃ ∼ N (0, 1), and z̃l ∼ N (0, σz2 )with σz2 > 0.
Supposing that random variances (ṽ, η̃, ε̃ i∈[0,1] , ũ, z̃l ∈[0,1] ) are mutually independent.
The total trading of active noise traders is
Z L
X̃ = x̃l dl = (ũ + qδ̃) L. (4)
0

Obviously, private information dissemination causes qδ̃L more noise trading. The
precise of X̃ is defined as ρ X = 1/Var ( X̃ ), then

1
ρX = . (5)
L2 (1 + q2 )

4. The Equilibrium
This study is based on the Grossman and Stiglitz [6] concept of REE. At the trading
time t = 1, informed traders make their trading to maximize their expected utility, and
senior noise traders and junior noise traders trade an amount of asset to maximize their
benefit net of expected trading loss. The demand functions of these three types of traders
are different. The following subsections present a discussion of the equilibrium.

4.1. Financial Market Equilibrium

Assuming that every trader believes that price function has the following form:

p̃ = α0 + αy ỹ + αv ṽ + α X X̃. (6)

All coefficients in the above price function are endogenous. Following the spirit of
Kyle [34], market liquidity, LIQ = 1/α X .
As informed traders are CARA, the demand function for a informed trader is

E(ṽ| p̃, ỹ, s̃i ) − p̃

D ( p̃, ỹ, s̃i ) = ,
γVar (ṽ| p̃, ỹ, s̃i )

where E(ṽ| p̃, ỹ, s̃i ) is an informed trader i’s conditional mean of ṽ, and Var (ṽ| p̃, ỹ, s̃i ) is
the variance.
Sustainability 2022, 14, 11624 6 of 19

Given public signal ỹ, the information in price equivalent to signal s̃ p is

p̃ − α0 − αy ỹ α
s̃ p = = ṽ + X X̃. (7)
αv αv

Suppose that the precise value of s̃ p is ρ p ; then, s̃ p ∼ N (ṽ, ρ p ), ρ p = (αv /α X )2 ρ X . ρ p

reflects how much private information can be inferred from the price; that is, it reflects the
level of price informativeness.
By Bayes’ rule, the following expression of E(ṽ| p̃, ỹ, s̃i ) and Var (ṽ| p̃, ỹ, s̃i ) can be
obtained:
ρη ỹ + ρ p s̃ p + ρε s̃i
E(ṽ| p̃, ỹ, s̃i ) = (8)
ρv + ρη + ρ p + ρε

and
1
Var (ṽ| p̃, ỹ, s̃i ) = . (9)
ρv + ρη + ρ p + ρε

Then, it is easy to obtain

ρη ỹ + ρ p s̃ p + ρε s̃i − (ρv + ρη + ρ p + ρε ) p̃
D ( p̃, ỹ, s̃i ) = . (10)
γ

The parameters ρ p , s̃ p , and p̃ are functions of q. In the market-clearing state, the

following condition holds:
Z 1
D ( p̃, ỹ, s̃i )di + X̃ = 1. (11)
0

In equilibrium, the expressions of α0 , αy , αv , α X have the following form:

γ ρη
α0 = − ρ , αy = ρ ,
ρv + ρη + ( γε )2 ρ X + ρε ρv + ρη + ( γε )2 ρ X + ρε
( ργε )2 ρ X + ρε γ+
ρε
γ ρX
αv = ρ , αX = ρε 2 . (12)
ρv + ρη + ( γε )2 ρ X + ρε ρv + ρη + ( γ ) ρ X + ρε

where ρ X is a function of q.

4.2. Price Volatility

To analyze how private information dissemination affects price volatility through
noise traders, we decompose it as Ozsoylev et al. [1]:

Var ( p) = Var F ( p) + Var N ( p). (13)

where Var F ( p) is the fundamental-driven volatility driven by the volatility of future payoff
ṽ, and Var N ( p) is the noise-driven volatility caused by noise traders. We obtain the
expressions as follows:

α2v
Var F ( p) =
ρv
1 ( γe )2 L2 (11+q2 ) + ρε
ρ

= ( )2 ,
ρv ρv + ρη + ( ργe )2 2 1 2 + ρε
L (1+ q )
Sustainability 2022, 14, 11624 7 of 19

1
α2y L2 (1+ q2 )
Var N ( p) = +
ρη ρX
ρe 1 2
(γ + γ L2 (1+ q2 ) ) 1
= (1 + 1
) .
L2 (1+ q2 )
(ρv + ρη + ( ργe )2 L2 (11+q2 ) + ρ ε )2

Obviously, Var F ( p) and Var N ( p) are related to private information dissemination q.

Theorem 1. Private information dissemination reduces noise-driven volatility but increases

∂Var F ( p)
fundamental-driven volatility. That is, ∂q > 0, ∂ ∂Var∂qN ( p) < 0.

The dissemination of private information diminishes the advantages of informed

traders with private information and creates greater fundamental-driven volatility. The
trading volume of informed traders will be decreased due to the weakening of the ad-
vantages of having private information; therefore, less fundamental value is reflected in
the market, leading to more fundamental-driven price volatility. The dissemination of
private information weakens noise-driven volatility. On the one hand, the dissemination of
private information favors senior noise traders. As private information is disseminated,
the proportion of senior noise traders in the market increases, and their trading decisions
become more rational than when they do not receive private information. On the other
hand, the private information dissemination increases information asymmetry and leads
to fewer noise traders being active in trading. As a result, the price volatility driven by
them decreases. Price volatility may adversely affect the sustainable development of the
financial market.

4.3. The Equilibrium under an Endogenous L

In this part, we analyze noise traders’ decisions. All noise traders participating in the
trading can acquire an exogenous liquidity gain, B > 0. The expected return of noise trader
l is E[(ṽ − p̃) x̃l ]. A positive x̃l means that noise trader l buys x̃l units of a risky asset at a
price p̃ at t = 2 and then sells them at t = 3 with a payoff ṽ.
Similarly, a negative x̃l means that noise trader l shorts x̃l units of a risky asset and
obtains x̃l p̃ payoffs at t = 2, and then produces an obligation of x̃l ṽ at t = 3.
For a given number of noise traders L, the expected utility of a noise trader is

W ( L; q) = B + E[(ṽ − p̃) E[ x̃l ]] = B − (1 + q2 )α x ( L; q) L. (14)

The equilibrium L∗ is defined by

W ( L∗ ; q) = 0. (15)

It is easy to prove that W (0; q) > 0, lim W ( L; q) = −∞. As W ( L; q) is a continuous

L→∞
function, by the intermediate value theorem, there exists at least one equilibrium L∗ . Since
∂α x ( L; q)/∂L < 0 and ∂α x ( L; q)/∂q < 0, it is obvious that L affects W ( L; q) in two opposite
ways, and q affects W ( L; q) positively.
As LIQ = 1/α x and ∂α x ( L; q)/∂q < 0, this shows that a larger q leads to a
stronger liquidity.

Theorem 2. (i) For any given (γ; ρv ; ρη ; B; ρε ; q) ∈ R6+ , there exists an equilibrium L∗ , where L∗
is determined by
ρε 1
γ+ γ L ∗2 (1+ q2 )
∗
W ( L ; q ) = B − (1 + q 2
) L∗ = 0. (16)
ρv + ρη + ( γε )2 L∗2 (11+q2 )
ρ
+ ρε
Sustainability 2022, 14, 11624 8 of 19

3ρε 3(1+q2 )ρε −[1−2(1+q2 )2 ](ρv +ρη +ρε )

(ii) If (1+q2 )(ρv +ρη +ρε )2
≤ B2 ≤ 3(1+q2 )(ρv +ρη +ρε )2
and 8B6 (1 + q2 )3 [1 − 2(1 +
q2 )2 ](ρv + ρη + ρε )5 − 4B4 (1 + q2 )3 [18B2 (1 + q2 )ρε + 1](ρv + ρη + ρε )4 − 30B4 (1 + q2 )2 [1 −
2(1 + q2 )2 ](ρv + ρη + ρε )3 ρε 3B2 [1 − 2(1 + q2 )2 ](ρv + ρη + ρε )2 ρε + 18B2 (1 + q2 )[1 − 2(1 +
q2 )2 ](ρv + ρη + ρε )ρε − [27B2 (1 + q2 )ρε + 9(1 + q2 )2 + 1](1 + q2 )ρ2ε > 0, there exist three
equilibria L∗ ; otherwise, there is a unique equilibrium except for a parameter set of Lebesgue
measure zero.

Proof. See Appendix A.

4.4. The Property of the Equilibrium

Following Crego [22], an equilibrium L∗ is stable when ∂W ( L∗ ; q)/∂L < 0, and un-
stable when ∂W ( L∗ ; q)/∂L > 0. Since W (0; q) > 0 and lim W ( L; q) = −∞, if there is a
L→∞
unique equilibrium, it must be a stable one, and if there are three equilibria, denoted by
L1∗ , L2∗ , L3∗ , ( L1∗ < L2∗ < L3∗ ), then W ( L; q) has two extreme points: the one between L1∗ and
L2∗ is negative, and the other one between L2∗ and L3∗ is positive. With the definition of
stability above, it is obvious that L1∗ and L3∗ are stable and that L2∗ is unstable.

4.5. Implication of Information Dissemination

We discuss the influence of private information dissemination on the market liquidity
and price efficiency in this part. Private information dissemination increases the information
asymmetry. Some active noise traders will choose to be inactive. The faster the private
information spread, the more noise traders will exit the market, i.e., ∂L/∂q < 0. The
existing active noise traders who receive the disseminate information will adjust their
trading volumes. Good (bad) private information dissemination helps senior noise traders
to increase (decrease) their trading volumes. We know that an equilibrium L∗ is stable if
and only if ∂W ( L∗ ; q)/∂L < 0. From (12), it is easy to obtain that

∂W ( L∗ ; q)/∂q = −(1 + q2 ) L∗ ∂α X ( L∗ ; q)/∂q + 2qL∗ α X ( L∗ ; q)

ρ2ε 2 2qρ3ε
2qγ[ρv + ρη + ρε )ρε +
γ2 L ∗ 2 (1+ q2 )
] + γL∗2qρ ε
(1+ q2 )
+ γ3 L ∗ 2 (1+ q2 )2
=− .
ρ2ε
[ρv + ρη + ρε )ρε + ∗ 2
γ L (1+ q2 )
2 ]2

For any q ∈ (0, 1), ∂W ( L∗ ; q)/∂q < 0 always holds. Using an implicit theorem for (12),
∂W ( L∗ ;q) ∂W ( L∗ ;q)
it can be obtained that ∂L∗ /∂q = − ∂q / ∂L < 0.
The liquidity LIQ is a function of ( L∗ ; q), where L∗ is with respect to q, so

∂LIQ( L∗ ; q) ∂LIQ( L∗ ; q) ∂L∗ ∂LIQ( L∗ ; q)

= + . (17)
∂q ∂L ∂q ∂q

The first term is an indirect effect, and the second is a direct effect. Since LIQ( L∗ ; q) =
1/α X ( L∗ ; q), ∂α X ( L∗ ; q)/∂q < 0, ∂α X ( L∗ ; q)/∂L < 0, ∂L∗ /∂q < 0, we find that the neg-
ative impact of the first term outweighs the positive impact of the second term in (17).
Thus, private information dissemination affects the market liquidity negatively. Private
information dissemination leads to a smaller volume of noise trading, which hinders the
market liquidity. The price efficiency decreases with private information dissemination. As
private information dissemination results in a lower proportion of informed trading, less
fundamental information is revealed in price. With the expression ρ p = (αv /α X )2 ρ X , then

ρε 2 1
ρp = ( ) 2 . (18)
γ L (1 + q2 )

By the implicit theorem,

Sustainability 2022, 14, 11624 9 of 19

∂W ( L∗ ; q) ∂W ( L∗ ; q)
∂ρ∗p /∂q = − / .
∂q ∂ρ p

γ(ρv + ρη )
∂W ( L∗ ; q)/∂ρ p = −(1 + q2 ) L∗ .
ρ2ε
ρε [ρv + ρη + ρε )ρε +
γ2 L ∗ 2 (1+ q2 )
]2

Thus, ∂ρ∗p /∂q < 0, information dissemination has a negative effect on price efficiency.

Theorem 3. Private information dissemination increases the information asymmetry and reduces
∗
the the number of noise traders; formally, ∂L
∂q < 0. The market liquidity is decreased through private
∂LIQ
information dissemination; formally, ∂q < 0. The price efficient is reduced by private information
∂ρ∗p
dissemination; formally, ∂q < 0.

The work by Han et al. [8] is a special case in our study. This paper generalizes the
research of the literature.

5. Endogenous Information Acquisition

Information acquisition is taken into account in this section. Assume that informed
trader i acquires private information at t = 0 with a cost function C (ρεi ) ≥ 0. C (ρεi ) is a
monotonous increase and convex.

5.1. The Equilibrium

Suppose that informed trader i believes the precisions of other informed trader’ private
signals are ρ∗ε . Just like Grossman and Stiglitz [6], informed trader i’s expected net benefit
of the private signals is

1 Var (ṽ|ỹ, p̃)

π (ρεi ; ρ∗ε ; L∗ ) = log[ ] − C (ρεi ). (19)
2γ Var (ṽ|ỹ, p̃, s̃i )

The informed trader i is arbitrary above, so ρεi = ρ∗ε is established. Using the first
order condition ∂π (ρεi ; ρ∗ε ; L∗ )/∂ρεi = 0, the following equation can be obtained via direct
calculation:
1
ρ∗
= C 0 (ρ∗ε ). (20)
2γ[ρv + ρη + ( γε )2 L∗2 (11+q2 ) + ρ∗ε ]

ρ∗ε is implicitly determined, ρ∗ε = ρε ( L∗ ; ρη ; q). Distinctly, the equilibrium condition

for L∗ is
ρ∗ε 1
γ+ γ L ∗2 (1+ q2 )
B− ρ∗
L∗ = 0, (21)
ρv + ρη + ( γε )2 L∗2 (11+q2 ) + ρ∗ε

where L∗ is implicitly determined; L∗ = L(ρ∗ε ; ρη ; q).

Theorem 4. There exists an overall equilibrium where the precision ρ∗ε ∈ (0, ρ̄ε ) of informed
traders’ private information is determined by

1
B2 (1 + q2 )[ − (ρv + ρη + ρ∗ε )] = [1 − 2γC 0 (ρ∗ε )(ρv + ρη )]2 , (22)
2γC 0 (ρ∗ε )

where ρ̄ε is uniquely determined by

Sustainability 2022, 14, 11624 10 of 19

1
= ρv + ρη + ρ̄ε , (23)
2γC 0 (ρ̄ε )

the equilibrium mass L∗ of noise traders who choose to take part in the market is

ρ∗ε
L∗ = r (24)
1+ q2
γ 2γC 0 (ρ∗ε )
− (1 + q2 )(ρv + ρη + ρ∗ε )

The equilibrium price function is p̃ = α0 + αy ỹ + αv ṽ + α X X̃, with ρ X = 1/L∗2 (1 + q2 ). If

q
B > 27(ρ +ρ8 )(1+q2 ) , the overall equilibrium is unique.
v η

Proof. See Appendix B.

5.2. The Effect of Information Dissemination

For price efficiency ρ∗p , the following expression can be obtained via direct calculation:

dρ∗p C 00 (ρ∗ε ) dρ∗ε

= −( + 1 ) , (25)
dq 2γC 0 (ρ∗ε ) dq

dρ∗ε
where C 0 (ρ∗ε ) > 0, C 00 (ρ∗ε ) > 0 (c(ρε ) is a convex function). > 0 when
q dq
B > 27(ρ +ρ8 )(1+q2 ) .
v η

q
8
Theorem 5. Assuming that B > 27(ρv +ρη )(1+q2 )
. In equilibrium with endogenous information
dρ∗p
acquisition, private information dissemination weakens price efficiency. Formally, < 0 in the
dq
overall equilibrium.

Proof. See Appendix C.

Informed traders are the main contributors to price efficiency. The dissemination of
private information to noise traders will seriously dilute the information advantages of
informed traders, leading to a significant reduction in their trading volume. This negative
effect overcomes the positive effect caused by a decrease in the number of noise traders.
Thus, the dissemination of private information reduces price efficiency.

6. Numerical Results
In this section, several economic situations are analyzed to intuitively observe the
influence of information dissemination on the number of equilibrium noise traders, market
liquidity, and price efficiency. The parameter q controls the speed of information dissemi-
nation. Following Han et al. [8] , we define that C (ρε ) = 2c ρ2ε , where c > 0 is constant. Set
γ = 2, ρv = 1.
Figure 1 displays the trend of the equilibrium noise traders’ number, market liquidity,
and price efficiency with information dissemination in economics with different information
costs. The number of noise traders L decreases with the increase of information dissemina-
tion q in Figure 1a. According to economic theory, individuals’ behavioral decisions are
more inclined to deterministic results and are averse to uncertain results. Information dis-
semination increases information asymmetry among traders and leads fewer junior noise
traders to participate in the market trading. We know that the information cost coefficient c
may affect informed traders’ information acquisition. In order to observe whether changes
in the value of c will cause different results, we observe the three curves of the equilibrium
values of the noise traders number L in Figure 1a. We find that the trend of the equilibrium
noise traders’ numbers decrease with the information costs c. In an economy with a higher
Sustainability 2022, 14, 11624 11 of 19

information cost, the advantages of information become more apparent. Moreover, with
information dissemination, more senior noise traders become active. However, when the
information disseminates more widely, the information advantage of senior noise traders is
weakened; thus, a faster decreasing rate of the equilibrium noise traders arises.

c=0.2
0.25 5
c=0.5 0.5
c=0.2
c=0.8 c=0.2
0.2 4 c=0.5
0.45 c=0.5
c=0.8
c=0.8
0.15 0.4 3

ρp
LIQ
L

0.35
0.1 2
0.3
0.05 1
0.25

0 0.2 0
0 0.2 0.4 0.6 0.8 1 0 0.2 0.4 0.6 0.8 1 0 0.2 0.4 0.6 0.8 1
q q q
(a) (b) (c)

Figure 1. (a): the effect of information dissemination on the number of equilibrium noise traders in
economics. (b): the effect of information dissemination on market liquidity in economics. (c): the the
effect of information dissemination on price efficiency in economics. Information acquisition costs in
the three economics are set: c = 0.2, c = 0.5, c = 0.8. Parameter ρη = 1.

Because the information cost affects informed traders’ information acquisition, it

may make an impact on the market liquidity. Figure 1b is the the equilibrium values of
market liquidity with an increase of information dissemination in situations where the
information acquisition costs are different: c = 0.2, c = 0.5, c = 0.8. It is obvious that the
market liquidity decreases with information dissemination for all of the three situations.
Comparing the situation where c = 0.2 and c = 0.5, we find that the equilibrium market
liquidity increases with the information costs c. When the information cost is higher, the
advantages of traders who know the private information are highlighted. The information
advantage attracts traders who have the private information to trade more, which leads
to better market liquidity. Furthermore, the advantages become stronger with a higher
information cost. From a comprehensive comparison of all curves in Figure 1b, we obtain
the same conclusion.
Figure 1c plots the influence of information dissemination on price efficiency. In
consideration of the effects of the information cost coefficient c on information acquisition,
we plot three cases with different information cost values to make a comparative analysis. It
shows that price efficiency ρ p decreases with q. By analyzing the three curves, we discover
that the higher the information cost, the lower the price efficiency. On the one hand, when
the information cost is higher, the informed trading volume will be reduced. Therefore, less
fundamental information is reflected in the price. On the other hand, higher information
costs reduce the information acquisition, which leads to weaker information dissemination
impacts. Thus, a higher information cost results in lower price efficiency.
Han et al. [8] study the influence of public information on the number of noise traders,
market liquidity, and price efficiency, but they have not considered information dissem-
ination. Figure 2 shows the effect of public information on the number of noise traders,
market liquidity, and price efficiency under different degrees of information dissemination,
such as the case where q = 0 equals the situation in Han et al. [8] In Figure 2a, private
information dissemination weakens the effect of public information on the number of noise
traders. Figure 2b displays that private information dissemination weakens the impact of
public information market liquidity. The reinforcement increases with public information
precision. Thus, there is a significant overestimation in Han et al. [8] In Figure 2c, the
effect of public information on price efficiency is enhanced by information dissemination,
and this effect decreases with the information dissemination. Therefore, effects of public
information on noise traders’ number and market liquidity are overestimated and that on
price efficiency is underestimated in Han et al. [8].
Sustainability 2022, 14, 11624 12 of 19

0.04 0.5 25
q=0 q=0 q=0
q=0.5 q=0.5 q=0.5
20
0.03 q=1 0.4 q=1 q=1

15

LIQ

ρp
0.02 0.3

L
10

0.01 0.2
5

0 0.1 0
0 2 4 6 8 10 0 2 4 6 8 10 0 2 4 6 8 10
ρη ρη ρη
(a) (b) (c)

Figure 2. (a): the effect effect of public information on the equilibrium number of noise traders; (b):
the effect effect of public information on the equilibrium number of market liquidity; (c): the The
effect effect of public information on the equilibrium number of price efficiency. The information
dissemination speeds are set: q = 0, q = 0.5, q = 1. The information acquisition cost parameter is
c = 0.2.

7. Conclusions
Information is a significant issue in financial market research. In this paper, we
take private information dissemination into account and establish a REE model to study
its impacts on price volatility, the number of noise traders, market liquidity, and price
efficiency. The results show that private information dissemination expands fundamental-
driven volatility, cuts down the number of noise traders in the market, reduces market
liquidity, and decreases price efficiency. Moreover, when extending our economy to the
situation where information is acquired endogenously, the same results are obtained.
This paper contributes to the analysis of the sustainability of financial markets. The
numerical results verify that the parameters’ value can affect the size but not the direction
or trend that the impact of information dissemination has on the financial markets. It can
be seen from the figures that an increase in information cost strengthens the influence of
information dissemination on noise traders’ number and market liquidity and weakens
the impact of information dissemination on the price efficiency. The number of noise
traders, market liquidity, and price efficiency all decrease with the increase in information
dissemination. In addition, we find that information dissemination negatively affects the
impact of public information on the number of noise traders and market liquidity but
positively affects the public information price efficiency.
The dissemination of private information to noise traders plays a role as a driver of the
non-sustainability of the financial market. Regulators should strengthen the supervision
of information disclosure, provide traders with continuous fundamental information to
weaken the negative impact of private information dissemination on the financial market,
and contribute to the sustainable development of the financial market.
We would also like to point out that this topic needs further research. First, the
heterogeneity of the noise traders is not considered in this study. In practice, the probability
of receiving private information varies with the social networks of noise traders. Second,
further empirical analysis is needed.

Author Contributions: Conceptualization, H.L. and W.L.; Methodology, H.L.; Writing—original

draft, H.L.; Writing—review & editing, Y.L. All authors have read and agreed to the published
version of the manuscript.
Funding: This research was funded by The Shanxi Province youth talent support program (grant
number 20190705); Humanities and Social Sciences Foundation of Ministry of Education, China
(grant number 20YJA910004); National Statistical Science Research Project of China (grant number
2022LZ14); Scientific Activities of Selected Returned Overseas Professionals in Shanxi Province (grant
number 20220025); National Natural Science Foundation of China (grant number 71971131).
Conflicts of Interest: The authors declare no conflict of interest.
Sustainability 2022, 14, 11624 13 of 19

Appendix A. Proof of Theorem 1

The equilibrium expected profit of a noise trader is

W ( L; q) = B − E[(ṽ − p̃) E[ x̃l ]] (A1)

= B − (1 + q2 )α X ( L; q) L (A2)
γ + γε L2 (11+q2 )
ρ

= B − (1 + q2 ) . (A3)
ρv + ρη + ( γε )2 L2 (11+q2 )
ρ
+ ρε

W ( L∗ ; q) = 0 is equivalent to
ρε ∗ ρε
γ(1 + q2 )( L∗ )3 − B(1 + q2 )(ρv + ρη + ρε )( L∗ )2 + L − B( )2 = 0. (A4)
γ γ

Let
ρε ρε
F ( L) = γ(1 + q2 ) L3 − B(1 + q2 )(ρv + ρη + ρε ) L2 + L − B ( )2 . (A5)
γ γ

Then, F ( L∗ ) = 0.
We can easily obtain
ρε
F 0 ( L) = 3γ(1 + q2 ) L2 − 2B(1 + q2 )(ρv + ρη + ρε ) L + . (A6)
γ

The discriminant of F 0 ( L) = 0 is

∆ = 4(1 + q2 )[ B2 (1 + q2 )(ρv + ρη + ρε )2 − 3ρε ], (A7)

3ρε
If ∆ ≤ 0, then F 0 ( L) > 0 for any L > 0. Thus, when B2 ≤ (1+q2 )(ρv +ρη +ρε )2
, F ( L)
monotonically increases with L. Thus, there exists a unique solution to (15). If ∆ > 0, then
3ρ
B2 > (1+q2 )(ρ +ε ρ +ρ )2 . In this case, there are two positive solutions to F 0 ( L) = 0, which are
v η ε
denoted by Ls and Lb :
ρε
Ls = q , (A8)
γ[ B(1 + q2 )(ρv + ρη + ρε ) + (1 + q2 )[ B2 (1 + q2 )(ρv + ρη + ρε )2 − 3ρε ]]

ρε
Lb = q . (A9)
γ[ B(1 + q2 )(ρ v + ρη + ρε ) − (1 + q2 )[ B2 (1 + q2 )(ρv + ρη + ρε )2 − 3ρε ]]

Using direct calculation, we obtain

F 00 ( L) = 6γ(1 + q2 ) L − B(ρv + ρη + ρε ). (A10)

F 00 ( Ls ) = 6γ(1 + q2 ) Ls − B(ρv + ρη + ρε ) (A11)

q
−2(1 + q2 ) [ B2 (1 + q2 )(ρv + ρη + ρε )2
= p q . (A12)
γ[ B 1 + q2 (ρv + ρη + ρε ) + B2 (1 + q2 )(ρv + ρη + ρε )2 − 3ρε ]
Sustainability 2022, 14, 11624 14 of 19

F 00 ( Lb ) = 6γ(1 + q2 ) Ls − B(ρv + ρη + ρε ) (A13)

q
2(1 + q2 ) [ B2 (1 + q2 )(ρv + ρη + ρε )2
= p q . (A14)
γ[ B 1 + q2 (ρv + ρη + ρε ) + B2 (1 + q2 )(ρv + ρη + ρε )2 − 3ρε ]

As F 00 ( Ls ) < 0 and F 00 ( Lb ) > 0, F ( L) increases on (0, Ls ) ∪ ( Lb , +∞), and decreases on

( Ls , Lb ). There are two equilibria L∗ if one of F ( Ls ) and F ( Lb ) is zero. Otherwise, there will
be a unique equilibrium L∗ . F ( Ls ) > 0 is equivalent to

B[4B2 (1 + q2 )2 (ρv + ρη + ρε )2 − 3(1 + q2 )ρε + [1 − 2(1 + q2 )2 ](ρv + ρη + ρε )]

q
B2 (1 + q2 )2 ( ρ v + ρ η + ρ ε )2 − 3(1 + q2 ) ρ ε
(A15)
< (1 + q2 )[9B2 (1 + q2 )(ρv + ρη + ρε )ρε + 2B2 (1 + q2 )2 (ρv + ρη + ρε )2
− 4B4 (1 + q2 )2 (ρv + ρη + ρε )3 − B2 (ρv + ρη + ρε )2 + 3(1 + q2 )ρε + ρε ]
F ( Lb ) < 0 is equivalent to

B[4B2 (1 + q2 )2 (ρv + ρη + ρε )2 − 3(1 + q2 )ρε + [1 − 2(1 + q2 )2 ](ρv + ρη + ρε )]

q
B2 (1 + q2 )2 ( ρ v + ρ η + ρ ε )2 − 3(1 + q2 ) ρ ε
(A16)
> (1 + q2 )[9B2 (1 + q2 )(ρv + ρη + ρε )ρε + 2B2 (1 + q2 )2 (ρv + ρη + ρε )2
− 4B4 (1 + q2 )2 (ρv + ρη + ρε )3 − B2 (ρv + ρη + ρε )2 + 3(1 + q2 )ρε + ρε ]
The two above inequalities are equivalent to B2 [4B2 (1 + q2 )2 (ρv + ρη + ρε )2 − 3(1 +
q2 )ρε + [1 − 2(1 + q2 )2 ](ρv + ρη + ρε )]2 [ B2 (1 + q2 )2 (ρv + ρη + ρε )2 − 3(1 + q2 )ρε ] < (1 +
q2 )2 [9B2 (1 + q2 )(ρv + ρη + ρε )ρε + 2B2 (1 + q2 )2 (ρv + ρη + ρε )2 − 4B4 (1 + q2 )2 (ρv + ρη +
ρε )3 − B2 (ρv + ρη + ρε )2 + 3(1 + q2 )ρε + ρε ]2 and B2 (1 + q2 )2 (ρv + ρη + ρε )2 − 3(1 + q2 )ρε >
0.
3ρε 3(1+q2 )ρε −[1−2(1+q2 )2 ](ρv +ρη +ρε )
Thus, if (1+q2 )(ρv +ρη +ρε )2
≤ B2 ≤ 3(1+q2 )(ρv +ρη +ρε )2
and 8B6 (1 + q2 )3 [1 −
2(1 + q2 )2 ](ρv + ρη + ρε )5 − 4B4 (1 + q2 )3 [18B2 (1 + q2 )ρε + 1](ρv + ρη + ρε )4 − 30B4 (1 +
q2 )2 [1 − 2(1 + q2 )2 ](ρv + ρη + ρε )3 ρε 3B2 [1 − 2(1 + q2 )2 ](ρv + ρη + ρε )2 ρε + 18B2 (1 + q2 )[1 −
2(1 + q2 )2 ](ρv + ρη + ρε )ρε − [27B2 (1 + q2 )ρε + 9(1 + q2 )2 + 1](1 + q2 )ρ2ε > 0, there exist
three equilibria L∗ .
3ρε 3(1+q2 )ρε −[1−2(1+q2 )2 ](ρv +ρη +ρε )
Similarly, we can prove that if (1+q2 )(ρv +ρη +ρε )2
≤ B2 ≤ 3(1+q2 )(ρv +ρη +ρε )2
and 8B (1 + q ) [1 − 2(1 + q ) ](ρv + ρη + ρε ) − 4B (1 + q ) [18B2 (1 + q2 )ρε + 1](ρv +
6 2 3 2 2 5 4 2 3

ρη + ρε )4 − 30B4 (1 + q2 )2 [1 − 2(1 + q2 )2 ](ρv + ρη + ρε )3 ρε 3B2 [1 − 2(1 + q2 )2 ](ρv + ρη +

ρε )2 ρε + 18B2 (1 + q2 )[1 − 2(1 + q2 )2 ](ρv + ρη + ρε )ρε − [27B2 (1 + q2 )ρε + 9(1 + q2 )2 + 1](1 +
q2 )ρ2ε = 0, there will be two equilibria L∗ .
If the values of ( B, ρv + ρη , ρε , 1 + q2 ) are in the space R3 but none of the above two
conditions are satisfied, the equilibrium is unique.

Appendix B. Proof of Theorem 3

We know the equilibrium is determined by joining (20) and (21). Equation (20) can be
ρ∗ε
rewritten as the following: L∗ = r .
1+ q2
γ 2γC0 (ρ∗ ) − (1 + q2 )(ρv + ρη + ρ∗ε )
ε
Sustainability 2022, 14, 11624 15 of 19

Taking the expression of L∗ into (21), we have

1 − 2γC 0 (ρ∗ε )(ρv + ρη )

B= q . (A17)
(1 + q2 )[ 2γC10 (ρ∗ ) − (ρv + ρη + ρ∗ε )]
ε

Equation (A17) can be rewritten as

1
B2 (1 + q2 )[ − (ρv + ρη + ρ∗ε )] = [1 − 2γC 0 (ρ∗ε )(ρv + ρη )]2 . (A18)
2γC 0 (ρ∗ε )

Appendix B.1
Define

1
G (ρε ) = − ( ρ v + ρ η + ρ ε ), (A19)
2γC 0 (ρε )

then,

C 00 (ρε ) + 2γC 0 (ρε )

G 0 (ρε ) = − . (A20)
2γC 0 (ρε )

Since C (ρε ) is convex, we have C 0 (ρε ) > 0 and C 00 (ρε ) > 0. Thus, G 0 (ρε ) < 0 and
G (ρε ) is monotonous. Calculating directly, limρε →∞ G (ρε ) = −∞, G (0) = ∞, so there is a
unique ρ̄ε that satisfies C (ρ̄ε ) = 0. For all C (ρε ) > 0, we have ρε < ρ̄ε .

Appendix B.2
From the proof above, we can easily obtain that B2 (1 + q2 ) G (ρε ) is monotonous in
(0, ρ̄ε ). When ρε ∈ (0, ρ̄ε ), B2 (1 + q2 ) G (ρε ) ∈ (0, ∞), [1 − 2γC 0 (0)(ρv + ρη )] = 1. Use
Equation (23), [1 − 2γC 0 (ρ̄ε )(ρv + ρη )] = ρv +ρηε +ρ̄ε . When ρε ∈ (0, ρ̄ε ), [1 − 2γC 0 (ρε )(ρv +
ρ̄

ρη )] ∈ ( ρv +ρηε +ρ̄ε , 1). Under the intermediate value theorem, there is a ρ∗ε that satisfies
ρ̄

Equation (22).

Appendix B.3
Define

B2 (1 + q2 )
K(x) = − [1 − x (ρv + ρη )]2 , (A21)
x

where x = 2γC 0 (ρ∗ε ).

It is easy to acquire that

B2 (1 + q2 )
K0 (x) = − + 2(ρv + ρη )[1 − x (ρv + ρη )]. (A22)
x2

2x2 (ρv +ρη )[1− x (ρv +ρη )]

When K 0 ( x ) < 0, there is B2 > max{ (1+ q2 )
8
} =
27(1+q2 )(ρv +ρη )
. Thus,
when B2 > 8
27((ρv +ρη ))
, K0 (x) 8
< 0, K ( x ) is decreasing. When B2 > 27((ρv +ρη )) , B2 (1 +
q2 )(ρv + ρη + ρ∗ε ) is increasing in ρ∗ε . Using the intermediate theorem, there is a unique ρ∗ε
that satisfies the following equation:
Sustainability 2022, 14, 11624 16 of 19

B2 (1 + q2 )
− [1 − 2γC 0 (ρ∗ε )(ρv + ρη )]2 = B2 (1 + q2 )(ρv + ρη + ρ∗ε ). (A23)
2γC 0 (ρ∗ε )

Equation (A23) is equal to Equation (22).

Appendix C. Proof of Theorem 4

Appendix C.1
In the overall equilibrium, ρ∗ε = ρ∗ε ( L(ρ∗ε , q); q); thus,

dρ∗ε ∂ρ∗ ∂L∗ dρ∗ε ∂ρ∗ ∂L∗ ∂ρ∗

− ε = ε + ε. (A24)
dq ∂L ∂q dq ∂L ∂q ∂q

The above (A24) can be arranged as

dρ∗ε ∂ρ∗ ∂L∗ ∂ρ∗ ∂ρ∗ ∂L∗ −1

=( ε + ε )(1 − ε ) . (A25)
dq ∂L ∂q ∂q ∂L∗ ∂q

By the implicit function theorem, we obtain the following equations from Equation (20):

ρ∗ 2q
∂ρ∗ε ( γLε ∗ )2 (1+q2 )2
= ρ∗ ρ∗ε
. (A26)
∂q 2γC 00 (ρ∗ε )[ρv + ρη + ρ∗ε + ( γLε ∗ )2 1+1q2 ]2 + 1
(γL∗ )2 1+q2
+1

ρ∗
∂ρ∗ε ( γLε ∗ )2 L∗ (12+q2 )
= ρ∗ ρ∗ε
. (A27)
∂L 2γC 00 (ρ∗ε )[ρv + ρη + ρ∗ε + ( γLε ∗ )2 1+1q2 ]2 + 1
(γL∗ )2 1+q2
+1

ρ∗ 2q
γ (1+ q2 )2 ( ρ v + ρ η )
ε
∂L∗
= ρ∗ε ρ∗ε ∗ . (A28)
2 1 ∗ + ( ρ ε )2 1 )
∂q − γL ∗
1+ q 2 ( ρ v + ρ η ) + ( γ + γL ∗
1+ q 2 )( ρ v + ρ η + ρ ε γL ∗
1+ q 2

γ((ρ∗ε )2 +2( L∗ )2 (1+q2 )γ2 ρ∗ε +( L∗ )4 (1+q2 )2 γ4 −( L∗ )2 (1+q2 )γ2 ρη −( L∗ )2 (1+q2 )γ2 ρv )
∂L∗ ((ρ∗ε )2 +( L∗ )2 (1+q2 )γ2 ρv +( L∗ )2 (1+q2 )γ2 ρ∗ε +9L∗ )2 (1+q2 )γ2 ρv )2
= ρ∗ 2 (ρ +ρ ) ρ∗ ∗
1 )(ρ + ρ + ρ∗ +( ρε )2 1 )
. (A29)
∂ρε ε
γL∗ 1+q2 v η (γ+ γLε∗ v η γL∗
1+ q2 ε 1+ q2
− ∗ ρ∗
+ ρ∗
[ρv +ρη +ρε +( γLε∗ )2 1 2 ]2 ρv +ρη +ρ∗ε +( γLε∗ )2 1 2
1+ q 1+ q

∂ρ∗ε ∂L∗ ∂ρ∗ε

Since C (ρ∗ε ) is convex, C 00 (ρ∗ε ) > 0, the numerator of ∂L ∂q + ∂q > 0 is equivalent to
ρ∗ε 2 2 ∗ q ρ∗ ∗
( γL ∗) ( 1 ∂L
1+q2 L ∂q
+ 1+ q2
) > 0. Using Bayes’ rule, we have ∂ 1
∂q Var (ve|ye,e
p)
= −( γLε ∗ )2 1+2q2 ( L1 ∂L
∂q +
q
1+ q2
). Therefore,

∂ρ∗ε ∂L∗ ∂ρ∗ ∂ 1

+ ε =− . (A30)
∂L ∂q ∂q ∂q Var (ve|ye, pe)
Sustainability 2022, 14, 11624 17 of 19

∂ρ∗ε ∂L∗
By calculating directly, ∂ 1
∂q Var (ve|ye,e
p)
< 0 when ρ∗p > ρ∗ε . The denominator of ∂L ∂q +
∂ρ∗ε
∂q > 0 equals to

γ(ρ∗ε + ( L∗ )2 (1 + q2 )γ2 )
> 0. (A31)
(ρ∗ε )2 + ( L ∗ )2 (1 + q2 ) γ2 ρ ∗ 2 2 2 ∗ ∗ 2 2 2
v + ( L ) (1 + q ) γ ρ ε + ( L ) (1 + q ) γ ρ η

∂ρ∗ε ∂L∗
Thus, 1 − ∂L ∂ρ∗ε > 0 is always established.

Appendix C.2
Let

x = 2γC 0 (ρ∗ε ) ∈ [0, 2γC 0 (ρε )], (A32)

and

B2 (1 + q2 )
T (x) = − [1 − (ρv + ρη )]2 . (A33)
x

Obviously,

B2 (1 + q2 )
T 0 (x) = − + 2[1 − (ρv + ρη )](ρv + ρη ). (A34)
x2

By calculating the derivation of both sides of (25), we have

dρ∗ε 2qB2 (ρv + ρη + ρ∗ε − 1x )

= . (A35)
dq 2γC 00 (ρ∗ε ) T 0 ( x ) − B2 (1 + q2 )

Joining (23) and (A32), we obtain

1
ρv + ρη + ρ∗ε − <0 (A36)
x

always established.
Let T 0 ( x ) < 0; that is,

B2 (1 + q2 )
T 0 (x) = − + 2[1 − (ρv + ρη )](ρv + ρη ) < 0
x2
⇐⇒ B2 (1 + q2 ) > 2x2 [1 − (ρv + ρη )](ρv + ρη ). (A37)

It is easy to know that

max{2x2 [1 − (ρv + ρη )](ρv + ρη )| x ∈ [0, 2γC 0 (ρε )]}

2
= {2x2 [1 − (ρv + ρη )](ρv + ρη )| x = }
3( ρ v + ρ η )
8
= . (A38)
27(ρv + ρη )
Sustainability 2022, 14, 11624 18 of 19

Thus,

8
T 0 ( x ) < 0 ⇐⇒ B2 (1 + q2 ) > . (A39)
27(ρv + ρη )

dρ∗ε
q
8
Based on the above analysis, dq > 0 when B > 27(ρv +ρη )(1+q2 )
.

Appendix C.3
ρ∗
q
When B > 8
27(ρv +ρη )(1+q2 )
, as ρ∗p = ( γLε ∗ )2 1+1q2

1
ρ∗p = − (ρv + ρη + ρ∗ε ). (A40)
2γC 0 (ρ∗ε )

By calculating directly, we have

dρ∗p C 00 (ρ∗ε ) dρ∗ε

= −( ) . (A41)
dq 2γC 0 (ρ∗ε ) + 1 dq

dρ∗ε
Since C (ρε ) is convex, C 0 (ρ∗ε ) > 0 and C 00 (ρ∗ε ) > 0. Furthermore, dq > 0 when
q
B > 27(ρ +ρ8 )(1+q2 ) . Therefore, Theorem 4 is proven.
v η

References
1. Ozsoylev, H.N.; Walden, J.; Yavuz, M.D.; Bildik, R. Investor Networks in the Stock Market. Rev. Financ. Stud. 2013, 27, 1323–1366.
[CrossRef]
2. Ruan, L. Research on sustainable development of the stock market based on VIX index. Sustainability 2018, 10, 4113. [CrossRef]
3. Jeong, S.H.; Lee, H.S.; Nam, H.; Oh, K.J. Using a genetic algorithm to build a volume weighted average price model in a stock
market. Sustainability 2021, 13, 1011. [CrossRef]
4. Hur, S.K.; Chung, C.Y.; Liu, C. Is liquidity risk priced? Theory and evidence. Sustainability 2018, 10, 1809. [CrossRef]
5. Truong, T.T.T.; Kim, J. Premiums for non-sustainable and sustainable components of market volatility: Evidence from the Korean
stock market. Sustainability 2019, 11, 5123. [CrossRef]
6. Grossman, S.J.; Stiglitz, J.E. On the impossibility of informationally efficient markets. Am. Econ. Rev. 1980, 70, 393–408.
7. Han, B.; Yang, L. Social networks, information acquisition, and asset prices. Manag. Sci. 2013, 59, 1444–1457. [CrossRef]
8. Han, B.; Tang, Y.; Yang, L. Public information and uninformed trading: Implications for market liquidity and price efficiency.
J. Econ. Theory 2016, 163, 604–643. [CrossRef]
9. Hong, H.; Kubik, J.D.; Stein, J.C. Social Interaction and Stock-Market Participation. J. Financ. 2004, 59, 137–163. [CrossRef]
10. Ivkovi, Z.; Weisbenner, S. Information Diffusion Effects in Individual Investors’ Common Stock Purchases: Covet thy Neighbors’
Investment Choices. Rev. Financ. Stud. 2007, 20, 1327–1357. [CrossRef]
11. Cohen, L.; Frazzini, A.; Malloy, C. The small world of investing: Board connections and mutual fund returns. J. Political Econ.
2008, 116, 951–979. [CrossRef]
12. Pool, V.K.; Stoffman, N.; Yonker, S.E. The people in your neighborhood: Social interactions and mutual fund portfolios. J. Financ.
2015, 70, 2679–2732. [CrossRef]
13. Shiller, R.J. Comovements in stock prices and comovements in dividends. J. Financ. 1989, 44, 719–729. [CrossRef]
14. French, K.R.; Roll, R. Stock return variances: The arrival of information and the reaction of traders. J. Financ. Econ. 1986, 17, 5–26.
[CrossRef]
15. Xie, J.; Xia, W.; Gao, B. The sustainability of stock price fluctuations: Explanation from a recursive dynamic model. PLoS ONE
2021, 16, e0255081. [CrossRef] [PubMed]
16. Waygood, S. How do the capital markets undermine sustainable development? What can be done to correct this? J. Sustain.
Financ. Investig. 2011, 1, 81–87. [CrossRef]
Sustainability 2022, 14, 11624 19 of 19

17. Colla, P.; Mele, A. Information linkages and correlated trading. Rev. Financ. Stud. 2010, 23, 203–246. [CrossRef]
18. Ozsoylev, H.N.; Walden, J. Asset pricing in large information networks. J. Econ. Theory 2011, 146, 2252–2280. [CrossRef]
19. Halim, E.; Riyanto, Y.E.; Roy, N. Costly information acquisition, social networks, and asset prices: Experimental evidence.
J. Financ. 2019, 74, 1975–2010. [CrossRef]
20. Ahern, K.R. Information networks: Evidence from illegal insider trading tips. J. Financ. Econ. 2017, 125, 26–47. [CrossRef]
21. Van Ness, B.; Van Ness, R.; Yildiz, S. Private information in trades, R2, and large stock price movements. J. Bank. Financ. 2021,
131, 106194. [CrossRef]
22. Crego, J.A. Why does public news augment information asymmetries? J. Financ. Econ. 2020, 137, 72–89. [CrossRef]
23. Admati, A.R.; Pfleiderer, P. A Theory of intraday patterns: Volume and price variability. Rev. Financ. Stud. 1988, 1, 3–40.
[CrossRef]
24. Hayek, F.A. The use of knowledge in society. Knowl. Manag. Organ. Des. 1996, 35, 7–15.
25. Verrecchia, R.E. Information acquisition in a noisy rational expectations economy. Econometrica 1982, 50, 1415–1430. [CrossRef]
26. Duffie, D.; Manso, G. Information percolation in large markets. Am. Econ. Rev. 2007, 97, 203–209. [CrossRef]
27. Foerster, M. Dynamics of Strategic Information Transmission in Social Networks. Theor. Econ. 2019, 14, 253–295. [CrossRef]
28. Bowden, M.; McDonald, S. The Impact of interaction and social learning on aggregate expectations. Comput. Econ. 2008,
31, 289–306. [CrossRef]
29. Andrei, D.; Cujean, J. Information percolation, momentum and reversal. J. Financ. Econ. 2017, 123, 617–645. [CrossRef]
30. Black, F. Noise. J. Financ. 1986, 41, 528–543. [CrossRef]
31. Subrahmanyam, A. Risk aversion, market liquidity, and price efficiency. Rev. Financ. Stud. 1991, 4, 417–441. [CrossRef]
32. Bloomfield, R.; O’Hara, M.; Saar, G. How noise trading affects markets: An experimental analysis. Rev. Financ. Stud. 2009,
22, 2275–2302. [CrossRef]
33. Shefrin, H.; Statman, M. Behavioral capital asset pricing theory. J. Financ. Quant. Anal. 1994, 29, 323–349. [CrossRef]
34. Kyle, A.S. Continuous auctions and insider trading. Econometrica 1985, 53, 1315–1335. [CrossRef]
35. Dongwei, S. Noise trading and market quality in chinese stock markets. Econ. Res. J. 2008, 43, 82–95.
36. Chen Junhua, N.J. Stock index futures, noise trading and risk asset pricing. Rev. Investig. Stud. 2014, 33, 133–144.
37. Yuan, Y.; Qi, Z.; Jia, Z. Noise trading, momentum effect and momentum strategy. Manag. Rev. 2021, 33, 44–54.
38. Ben-Rephael, A.; Da, Z.; Israelsen, R.D. It depends on where you search: Institutional investor attention and underreaction to
news. Rev. Financ. Stud. 2017, 30, 3009–3047. [CrossRef]
39. Weiqi Liu, Y.R. Is institutional investor noise really small? Econ. Theory 2018, 37, 91–103.
40. Weiqi Liu, Y.D. Information advantage, limited attention and informed traders compete for trading. J. Appl. Stat. Manag. 2020,
39, 162–173.
41. Figlewski, S. Information diversity and market behavior. J. Financ. 1982, 37, 87–102. [CrossRef]
42. Waters, G.A. The many faces of the taylor rule for advanced undergraduate macroeconomics. Int. Rev. Econ. Educ. 2022, 41.
[CrossRef]
43. Zeng, Q.; Zhang, Q.; Liu, S.; Yang, Y. Illiquidity comovement and market crisis. J. Syst. Sci. Complex. 2022, 35, 1–12. [CrossRef]
[PubMed]
44. Wang, W.; Zhao, S.; Zhang, J. Multi-asset pricing modeling using holding-based networks in energy markets. Financ. Res. Lett.
2022, 46, 102483. [CrossRef]
45. Miao, J.; Su, D. Asset market equilibrium under rational inattention. Econ. Theory 2022, 32, 1–30. [CrossRef]
Reproduced with permission of copyright owner. Further reproduction
prohibited without permission.