Unveiling the Brew: Probing the Lingering Impact of the Luckin Coffee Scandal on the Liquidity of Chinese Cross-Listed Stocks

Abstract

This paper investigates the impact of the Luckin Coffee accounting scandal on stock liquidity and spillover effects in the financial market, focusing on Chinese companies listed on U.S. exchanges. Utilizing event studies, we analyze eight pivotal events related to the scandal to examine stock liquidity and market quality changes. The results show a significant decline in Luckin’s stock liquidity during the scandal, while spillover effects on other Chinese stocks are limited. Comparisons with the Satyam accounting scandal suggest that individual company scandals may not substantially affect the liquidity of other stocks from the same country. The findings highlight the importance of robust regulatory frameworks and investor due diligence in safeguarding market integrity and restoring investor confidence.

1. Introduction

In the world of high-speed growth and disruptive ambitions, few companies have soared—and fallen—as spectacularly as Luckin Coffee. Founded in October 2017, Luckin rapidly positioned itself as a leading coffee brand in China, using a low-cost, technology-driven approach to challenge established giants like Starbucks. Just nineteen months later, Luckin set a record as the fastest Chinese company to go public on NASDAQ (Leung 2019; Peng et al. 2022), raising $645 million in its May 2019 IPO. Yet, only eleven months after its IPO, in April 2020, the company admitted to fabricating approximately RMB 2.12 billion (USD 310 million) in sales. This revelation caused its stock price to plummet by over 80%, erasing nearly $5 billion in market capitalization (Zhu 2020). Investigations by U.S. and Chinese regulators followed, resulting in significant settlements and fines, and Luckin ultimately filed for Chapter 11 bankruptcy in 2021 to restructure its obligations. Though the company has since restructured (Lim 2024), its story remains a stark example of the risks posed by financial reporting irregularities among cross-listed firms.

While prior studies have examined the Luckin scandal from perspectives such as short-selling attacks (Peng et al. 2022) and corporate governance (Zhang and Obot 2024), its impact on the liquidity of Luckin’s stock and other cross-listed Chinese stocks remains unexplored. Our paper aims to fill this gap, motivated not only by the scandal itself but also by the decade-long concerns over fraudulent accounting and auditing practices among some Chinese firms (Darrough et al. 2020), which have triggered regulatory scrutiny and ongoing tensions between U.S. oversight bodies and their Chinese counterparts.

A brief examination can reveal that Luckin Coffee is not the only Chinese company that has perpetrated financial fraud. Companies like Yin Gungxia, Lan Tian, and Ke Long committed financial fraud, leading to extravagant losses for investors and creditors (Li and Wu 2007; Yang et al. 2017). One can always argue such cases as idiosyncratic, but it is undeniable that Chinese authorities restrict access to audit work, citing sovereignty (Wu 2022). While auditors of Chinese cross-listed firms in the U.S. market must register with the Public Company Oversight Board (PCAOB) and are subject to PCAOB inspections, the POAOB’s ability to inspect audit work for Chinese cross-listed companies is minimal (PCAOB 2021). Chinese cross-listed firms, therefore, remain shielded from full U.S. and Chinese regulatory oversight, which poses significant barriers to effective public oversight and audit scrutiny (Lamoreaux 2016). Consequently, it is hard for U.S. investors to detect fraudulent activities in time, as evidenced by the Luckin Coffee scandal.

Our research is further motivated by the broader issue of information asymmetry exacerbated by disparities in regulatory enforcement—one of the key determinants of stock liquidity. In this study, we focus on two related inquiries. First, we investigate the impact of the Luckin Coffee scandal on its stock liquidity, examining how significant corporate misbehavior influences trading behaviors and market quality. Unlike well-studied cases of U.S.-based companies like Enron, WorldCom, and Lehman Brothers (Akhigbe et al. 2005; Sadka 2006; Dumontaux and Pop 2013), the Luckin Coffee case offers a distinctive perspective as a Chinese-based firm cross-listed in the U.S., navigating a regulatory framework that diverges significantly from U.S. standards. This unique context underscores the significance of studying Luckin Coffee’s financial fraud, as it reveals the vulnerabilities of cross-border investments and highlights the risks for investors in firms operating within dual regulatory constraints.

Second, we analyze potential spillover effects on other Chinese cross-listed stocks, shedding light on the ripple effects of financial misconduct and regulatory opacity across the market. By exploring these dynamics, our study contributes to a deeper understanding of how trust, regulatory oversight, and information transparency affect liquidity not only for the company directly involved but also for other firms perceived as sharing similar risks. This research has broader implications for investors and regulators, offering insights into the unique challenges of ensuring market integrity in a globally interconnected financial environment.

The Luckin Coffee scandal exemplifies the complexities of agency conflicts (Clayman et al. 2012) in environments of high information asymmetry. While the initial inflation of revenues by Luckin’s management led to a surge in stock prices—seemingly benefiting shareholders—this short-term gain primarily served the interests of insiders, who gained more directly from the inflated valuation. Such actions highlight the misalignment between long-term shareholder interests and the priorities of company executives focused on short-term stock performance. We expect that the eventual exposure of this misalignment, alongside the revelation of financial manipulation, had a marked impact on stock liquidity. Following the scandal, investor confidence plummeted due to increased uncertainty about the company’s financial health, leading many to sell or avoid the stock, disrupting its demand–supply balance. Additionally, regulatory investigation and increased information asymmetry heightened perceived trading risks, while institutional investors withdrew to manage compliance and risk. All of these factors, therefore, jointly lowered the stock’s liquidity.

Investors, faced with a breach of trust, likely re-evaluated their risk assessments for other cross-listed Chinese firms operating under similar conditions of limited regulatory oversight. However, the potential spillover effect of the Luckin scandal on other Chinese cross-listed companies is more complex and may be influenced by a range of factors. While studies like Darrough et al. (2020) suggest that fraudulent behavior by Chinese firms can create a negative spillover effect on others due to country-of-origin bias and shifts in investor sentiment, it remains uncertain whether this dynamic applies directly to the Luckin case. On the one hand, non-fraudulent Chinese companies might experience reduced liquidity due to heightened risk perception and suspicion of similar misconduct (Donley et al. 2023; Peng et al. 2022; Wu 2022). Conversely, if investors interpret Luckin’s actions as idiosyncratic—specific to Luckin’s unique circumstances and management—then broader spillover effects might be minimal. Additionally, the timing of the Luckin scandal—coinciding with the global pandemic and heightened US–China trade tensions—may have tempered or complicated this effect, as investors were simultaneously navigating other significant macroeconomic uncertainties.

To address our inquiries, we conducted an event study revolving around eight pivotal events that shook Luckin Coffee to its core: (1) 7 January 2020: Luckin announces a follow-on offering of American depository receipts (ADRs) roughly nine months after its initial IPO in May 2019; (2) 14 January: Completion of the follow-on offering; (3) 31 January: The Muddy Waters Research report reveals financial irregularities; (4) 4 February: Ash Illumination Research exposes further financial misrepresentations; (5) 2 April: Luckin admits to overstating profits, leading to stock price declines and trading halts; (6) 21 May: NASDAQ announces its intention to delist Luckin; (7) 23 June: NASDAQ issues a second delisting announcement; and (8) 26 June: Luckin announces their delisting, which becomes effective on 29 June. Our pre-event periods are all trading days except for the specified event dates from January to June.

Our regression analysis of non-U.S. stocks traded on the New York Stock Exchange (NYSE) reveals a notable decline in Luckin’s stock liquidity alongside an increase in information-based trading and a decline in market quality throughout the event’s unfolding. Additionally, our results indicate no statistically significant spillover effect on other Chinese stocks. This suggests that the Luckin scandal may be perceived as an isolated incident rather than a systemic issue affecting Chinese stocks more broadly. For a robustness check of the spillover effects, we tested the stock liquidity using another scandal of Satyam Computer Services.¹ Notably, this lack of spillover effect aligns with findings from a similar event study examining the impact of the Satyam accounting scandal on Indian stocks traded on the NYSE in 2009. Thus, our research implies that the Luckin scandal did not exacerbate existing challenges other Chinese stocks face amidst the ongoing pandemic and prolonged trade tensions.

This study makes contributions to four areas of the literature. First, it enriches the understanding of the impact of accounting scandals on the stock market. While previous studies have primarily focused on how accounting fraud affects stock price formation, crash risk, and volatility (Griffin et al. 2004; Weske and Benuto 2015; Morris et al. 2019; Ahmad et al. 2021; Richardson et al. 2022), and returns (Beneish et al. 2012; Morris et al. 2019), our study breaks new ground by examining the influence of such a scandal on stock liquidity. This aspect is particularly important given the prior findings of Morris et al. (2019) on the long-term effects of stocks under investigation by the SEC for fraud. By examining the events surrounding the Luckin scandal and its aftermath, we offer valuable insight into the intricate relationship between corporate wrongdoing and market microstructure, particularly regarding stock liquidity and information-based trading.

Second, our research provides new empirical evidence of the repercussions of the Luckin scandal from various aspects, including the company’s corporate governance (Zhang and Obot 2024), damage to investors (Wang 2020), market responses (Peng et al. 2022), and potential tools for prevention and detection (Chen 2022). Furthermore, while most accounting fraud and spillover effect studies concentrate on firms in the same industry (Beatty et al. 2013; Brown et al. 2018), our study examines the spillover effect of firms originating from the same country as those involved in accounting fraud. By investigating the absence of statistically significant spillover effects on other Chinese stocks traded on the NYSE in the case of Luckin and Indian stocks in the case of Satyam, we illuminate the isolated nature of such scandals and their implications for investor sentiment and market behavior. Finally, this study sheds light on how agency conflicts, compounded by information asymmetry, can destabilize investor confidence and affect market liquidity across interconnected stocks, offering a fresh perspective compared to accounting frauds involving U.S. domestic companies.

2. Data and Variables

From the Center for Research in Security Prices (CRSP) database, we identified 1,002 non-U.S. stocks listed on major U.S. exchanges and 145 stocks from China in 2020. Based on this list, we retrieved information on liquidity variables for the non-U.S. stocks using the NYSE’s Trade and Quote database (TAQ), which has comprehensive historical data for trading activity measures. Standard filters used in the market microstructure were applied to ensure data accuracy. These filters removed errors or anomalies, such as negative bid or ask prices/sizes, out-of-order trades/quotes, and those outside regular trading hours. Trades with negative prices/volumes and significant deviations exceeding 10% from the last transaction were also excluded. These measures enhanced data reliability, ensuring their suitability for subsequent analysis.

We measured liquidity and information asymmetry with four variables, calculated below:

$Quoted Sprea{d}_{j,t}=\left(As{k}_{j,t}-Bi{d}_{j,t}\right)/Mi{d}_{j,t}$

$Effective Sprea{d}_{j,t}=2D_{j,t}\left(P_{j,t}-Mi{d}_{j,t}\right)/Mi{d}_{j,t}$

$Realized Sprea{d}_{j,t}=2D_{j,t}\left(P_{j,t}-Mi{d}_{j,t+5}\right)/Mi{d}_{j,t}$

$Price Impac{t}_{j,t}=100D_{j,t}\left(M_{j,t+5}-Mi{d}_{j,t}\right)/Mi{d}_{j,t}$

where $As{k}_{j,t}$ is the ask price for stock j at time t, $Bi{d}_{j,t}$ is the bid price for stock j at time t, $Mi{d}_{j,t}$ is the mean of $As{k}_{j,t}$ and $Bi{d}_{j,t}$, $P_{j,t}$ is the transaction price for stock j at time t, and $D_{j,t}$ is a binary variable that equals one for customer buy orders and negative one for customer sell orders. We estimated $D_{j,t}$ using the algorithm in Ellis et al. (2000). Realized spread and price impact were also used as proxies to capture the information asymmetry. The descriptive statistics of the variables used are reported in Table 1.

3. Empirical Results

3.1. Results on Stock Liquidity

We employed the following OLS regression model to explore the relationship between liquidity and Luckin’s accounting scandal:

$\begin{array}{l}L{Q}_{it }=a+{\beta }_{1}EventDumm{y}_{it}+{\beta }_{2}Chin{a}_{it} or Luckin Dumm{y}_{it}+{\beta }_{3}Interactio{n}_{it}\\ +{\sum }_{l=2}^n{\beta }_l{X}_{it}^l+{\epsilon }_{ijt}\end{array}$

where LQ measures stock liquidity, including quoted spread, effective spread, realized spread, and price impact; i represents each stock; j represents each country to which stock i belongs; t represents time; and d denotes each event date related to the Luckin scandal. Realized spread and price impact were also used as proxies for information-based trading. Following the existing literature (Kim et al. 2024a, 2024b), we used the controls that can explain stock liquidity. X represents a set of stock-specific standard control variables, such as stock price, return volatility, and dollar trading volume, with firm i and time t; and ε represents the error term. We have included the description of the variables in Appendix A.

3.1.1. Results on Luckin Stock

Our event variable equaled 1 for any of the eight event dates and 0 for non-event dates from January to June 2020. Since the scandal occurred during the pandemic and prolonged trade tensions between the U.S. and China, we assumed that the effect of COVID-19 may play a significant role. To control for the impact of COVID-19, we used a fear index related to COVID-19 (created by Salisu and Akanni 2020) and a dummy variable for stocks from China to help isolate the effect of the Luckin scandal. The fear index was the equally weighted measure of the reported case index (RCI) and reported death index (RDI)², used by Salisu and Akanni (2020), Salisu et al. (2020), and Mazumder and Saha (2021). Therefore, our key variables of interest were the Luckin dummy and the interactive terms. We employed a heteroscedasticity-robust method with Huber–White estimators to estimate the equation.

Table 2 presents the regression results linking Luckin’s accounting fraud scandal to stock liquidity. The coefficients of the Luckin dummy are positive and statistically significant for all four liquidity measures, indicating declined liquidity for the entire sample period. The positive sign of the coefficients of the interactive term indicates that Luckin experienced more diminished liquidity when the scandal unfolded, although insignificant for realized spread. Additionally, the coefficients of the fear index are positive and significant, suggesting that COVID-related fear decreased the liquidity of all non-U.S. stocks.

3.1.2. Effects on Stock Liquidity by Event Dates

To explore the scandal’s nuanced effects further, we conducted a granular analysis by examining the impact of each event date individually. This approach allowed us to assess how stock liquidity was influenced at different points throughout the scandal’s unfolding. By introducing interactions between the Luckin dummy variable and each event dummy variable, we could discern whether the effects on liquidity varied depending on the specific event date.

Table 3 summarizes the results, revealing variations in the signs of the interactive terms. The interaction between the Luckin dummy and the first event, where Luckin announced the pricing of a follow-on offering of ADRs, is positive and significant. This suggests that investors expressed concern about the company’s new offering only nine months after its IPO. Similarly, the other interaction terms, except for event 7, are also positive and significant, indicating a decline in liquidity upon announcing these events. Event 7, which involved announcing a delisting notice from NASDAQ, stands out as the exception. For both the quoted and effective spreads, they are negative and statistically significant. This result suggests that the market might have perceived this information as positive, interpreting regulatory actions against scandal-involved stocks as a favorable signal.

3.2. Spillover Effect

Next, we examined whether the Luckin scandal had any spillover effect on other Chinese stocks traded in the U.S. Our analysis did not uncover any significant spillover effect. As shown in Table 4, although we observed wider spreads and a more significant price impact for Chinese firms on average, further examination of the interaction effect between the event dummy and the China dummy indicated negative (or zero) and statistically insignificant coefficients. This suggests that while there may have been some fluctuations in liquidity for Chinese stocks during the Luckin scandal, these effects were not consistent or significant enough to establish a clear spillover effect on the broader market.

As a robustness test, we also examined another accounting fraud case involving Satyam, an Indian IT services firm, to determine if spillover effects occurred. On 7 January 2009, Satyam confessed to manipulating revenues, margins, and cash balances amounting to INR 50 billion, which equated to approximately USD 1.02 billion at the prevailing exchange rate. Despite the severity of this accounting fraud incident, our analysis did not reveal any significant spillover effect onto other stocks from India listed on U.S. exchanges, as shown by the event dummy in Table 5. This finding provides additional evidence that one company scandal does not necessarily significantly affect the liquidity of other stocks from the same country.

4. Conclusions

This event study investigates the relationship between accounting scandals, stock liquidity, and spillover effects in the financial market. Our investigation focused on the Luckin Coffee scandal, a high-profile case that captivated global attention due to its audacious nature. We have uncovered valuable insights that may have important implications for various stakeholders in the financial markets.

For investors, particularly those interested in Chinese stocks or companies operating in similar contexts, the observed decline in liquidity associated with the Luckin scandal highlights the need for rigorous due diligence, especially for cross-listed stocks where transparency and regulatory scrutiny may be uneven. Investors should factor this into their decision making, as different regulatory standards can obscure financial risks. Moreover, the absence of a significant spillover effect on other Chinese stocks traded in the U.S. suggests that investors should not automatically extrapolate the implications of one scandal to the broader market. While individual events may have localized effects, the overall market may exhibit resilience in the face of such incidents.

Regulators play a critical role in maintaining market integrity and investor confidence. Our findings highlight the need for robust regulatory frameworks and enforcement mechanisms to prevent, detect, and mitigate the impact of accounting scandals. They should prioritize transparency, accountability, and investor protection, particularly in cross-border transactions and listings. The positive market reaction to regulatory actions, as evidenced by the delisting notice from NASDAQ, underscores the importance of swift and decisive regulatory intervention in restoring market confidence.

As Luckin’s case has shown, gaps in oversight create vulnerabilities that can be exploited, undermining investor trust and market integrity. Future research should focus on the role of regulatory disparities in shaping investor perceptions and behaviors toward cross-listed companies from different regions. This case also underscores the need for stronger standards to improve early detection and deterrence of potential fraud, as Luckin is unlikely to be the last high-profile scandal of its kind.

Author Contributions

Conceptualization, L.K., J.-C.K., S.M. and Q.S.; methodology, L.K., J.-C.K., S.M. and Q.S.; investigation, L.K., J.-C.K., S.M. and Q.S.; writing original draft, L.K., J.-C.K., S.M. and Q.S.; writing review and editing, L.K., J.-C.K., S.M. and Q.S.; visualization, L.K., J.-C.K., S.M. and Q.S. All authors have read and agreed to the published version of the manuscript.

Data Availability Statement

Data may be obtained from the corresponding author upon request.

Conflicts of Interest

The authors declare no conflicts of interest.

Footnotes

Notes

1. For many years, Satyam Computer Services inflated stock prices by reporting profits that never existed and by reporting cash at the bank that did not exist. They also fraudulently reported salary payments to reduce the taxable income. The significance of the event was very notable as analysts of India termed the Satyam scandal as India’s Enron scandal.

2. According to Salisu and Akanni (2020), the reported case index (RCI) measures how far people’s expectations on reported cases in the preceding 14-day period (incubation period) have veered from the present day’s reported cases. And the reported death index (RDI) measures how far peoples’ expectations from reported deaths in the preceding 14 days have veered from the present day’s reported deaths.

Footnotes

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Tables

Table 1

Descriptive statistics. This table reports the descriptive statistics of the variables used in this study. The descriptions of the variables are reported in Appendix A Table A1.

			Percentile
Variables	Mean	Standard Deviation	25th	50th	75th
Price ($)	23.74	51.86	2.91	8.82	24.01
Volatility (000)	0.0259	0.3568	0.0000	0.0003	0.0047
Volume ($000)	44,379	202,021	261	3,320	26,340
Quoted spread	0.0133	0.0263	0.0015	0.0043	0.0135
Effective spread	0.0091	0.0199	0.0009	0.0026	0.0091
Realized spread	0.0053	0.0195	0.0001	0.0007	0.0041
Price impact	0.0038	0.0161	0.0005	0.0013	0.0036
Fear index	41.59	30.47	0.00	46.94	52.15

Table 2

Regression results on Luckin stock. This table reports the regression results of the Lucking event and the interaction effect with the event dummy. Standard errors are adjusted for heteroscedasticity (Huber–White estimators). *** and ** indicate that the coefficients are statistically significant at 1% and 5% levels, respectively.

	(1)	(2)	(3)	(4)	(5)	(6)
Dependent Variables	(Quoted Spread)	(Quoted Spread)	(Quoted Spread)	(Effective Spread)	(Effective Spread)	(Effective Spread)
Event dummy	−0.0005 **		−0.0005 **	−0.0006 ***		−0.0006 ***
	(−2.10)		(−2.15)	(−4.10)		(−4.14)
LK dummy		0.0148 ***	0.0143 ***		0.0110 ***	0.0107 ***
		(23.55)	(22.75)		(25.87)	(24.87)
Event × K			0.0063 ***			0.0038 ***
			(2.97)			(2.87)
Fear Index	0.0001 ***	0.0001 ***	0.0001 ***	0.0001 ***	0.0001 ***	0.0001 ***
	(41.53)	(41.76)	(41.56)	(31.24)	(31.61)	(31.27)
Price	−0.0029 ***	−0.0029 ***	−0.0029 ***	−0.0006 ***	−0.0006 ***	−0.0006 ***
	(−23.89)	(−23.92)	(−23.92)	(−7.16)	(−7.19)	(−7.21)
Volatility	11.9623 ***	11.9584 ***	11.9570 ***	10.6795 ***	10.6775 ***	10.6756 ***
	(4.84)	(4.84)	(4.84)	(4.60)	(4.60)	(4.60)
Log(volume)	−0.0051 ***	−0.0051 ***	−0.0051 ***	−0.0036 ***	−0.0036 ***	−0.0036 ***
	(−97.21)	(−97.21)	(−97.21)	(−78.96)	(−78.96)	(−78.96)
Constant	0.0846 ***	0.0846 ***	0.0847 ***	0.0598 ***	0.0598 ***	0.0598 ***
	(102.15)	(102.21)	(102.15)	(82.01)	(82.10)	(82.01)
Observations	117,838	117,838	117,838	117,709	117,709	117,709
Adjusted2	0.4032	0.4034	0.4034	0.3802	0.3804	0.3804
	(1)	(2)	(3)	(4)	(5)	(6)
Dependent Variables	(Realized Spread)	(Realized Spread)	(Realized Spread)	(Price Impact)	(Price Impact)	(Price Impact)
Event dummy	−0.0002		−0.0002	−0.0006 ***		−0.0006 ***
	(−1.03)		(−1.04)	(−4.10)		(−4.14)
LK dummy		0.0081 ***	0.0079 ***		0.0110 ***	0.0107 ***
		(26.59)	(25.71)		(25.87)	(24.87)
Event × LK			0.0013			0.0038 ***
			(1.32)			(2.87)
Fear Index	0.0000 ***	0.0000 ***	0.0000 ***	0.0001 ***	0.0001 ***	0.0001 ***
	(16.89)	(16.92)	(16.90)	(31.24)	(31.61)	(31.27)
Price	−0.0010 ***	−0.0010 ***	−0.0010 ***	−0.0006 ***	−0.0006 ***	−0.0006 ***
	(−9.82)	(−9.85)	(−9.85)	(−7.16)	(−7.19)	(−7.21)
Volatility	5.6573 ***	5.6551 ***	5.6545 ***	10.6795 ***	10.6775 ***	10.6756 ***
	(4.15)	(4.15)	(4.15)	(4.60)	(4.60)	(4.60)
Log(volume)	−0.0026 ***	−0.0026 ***	−0.0026 ***	−0.0036 ***	−0.0036 ***	−0.0036 ***
	(−59.95)	(−59.94)	(−59.95)	(−78.96)	(−78.96)	(−78.96)
Constant	0.0418 ***	0.0418 ***	0.0418 ***	0.0598 ***	0.0598 ***	0.0598 ***
	(61.04)	(61.10)	(61.04)	(82.01)	(82.10)	(82.01)
Observations	117,697	117,697	117,697	117,709	117,709	117,709
Adjusted2	0.1800	0.1802	0.1802	0.3802	0.3804	0.3804

Table 3

Regression results summary on the interactive terms by event dates. This table reports the regression results of the interaction effect and the event dummies for each of the event separately. Standard errors are adjusted for heteroscedasticity (Huber–White estimators). ***, **, and * indicate that the coefficients are statistically significant at 1%, 5%, and 10% levels, respectively.

	(1)	(2)	(3)	(4)
Dependent Variables	(Quoted Spread)	(Effective Spread)	(Realized Spread)	(Price Impact)
Event1 dummy	0.0005	0.0002	0.0001	0.0004
	(1.04)	(0.53)	(0.16)	(1.12)
LK_dummy	0.0143 ***	0.0107 ***	0.0079 ***	0.0027 ***
	(22.41)	(24.61)	(25.57)	(9.54)
Event1 × LK	0.0065 ***	0.0043 ***	0.0027 ***	0.0015 ***
	(8.20)	(7.64)	(5.85)	(4.16)
Event2 dummy	0.0005	−0.0001	−0.0001	0.0003
	(1.02)	(−0.17)	(−0.43)	(1.36)
Event2 × LK	0.0066 ***	0.0047 ***	0.0032 ***	0.0013 ***
	(8.35)	(8.90)	(7.87)	(4.58)
Event3 dummy	0.0024 ***	0.0010 **	0.0016 **	−0.0004
	(4.13)	(2.55)	(2.16)	(−0.60)
Event3 × LK	0.0107 ***	0.0079 ***	0.0028 ***	0.0051 ***
	(12.65)	(13.82)	(3.44)	(7.09)
Event4 dummy	0.0016 ***	0.0004	−0.0007	0.0012 *
	(3.02)	(1.12)	(−1.00)	(1.77)
Event4 × LK	0.0068 ***	0.0051 ***	0.0046 ***	0.0005
	(8.39)	(9.19)	(6.00)	(0.77)
Event5 dummy	−0.0008	−0.0003	0.0001	−0.0005
	(−1.04)	(−0.45)	(0.25)	(−1.43)
Event5 × LK	0.0117 ***	0.0032 ***	−0.0026 ***	0.0058 ***
	(11.97)	(4.74)	(−4.35)	(14.68)
Event6 dummy	−0.0014 **	−0.0016 ***	−0.0005	−0.0010 ***
	(−2.49)	(−4.69)	(−1.28)	(−3.12)
Event6 × LK	0.0010	0.0012 **	−0.0006	0.0018 ***
	(1.15)	(2.25)	(−1.21)	(4.78)
Event7 dummy	−0.0035 ***	−0.0027 ***	−0.0014 ***	−0.0013 ***
	(−6.30)	(−8.07)	(−4.63)	(−6.32)
Event7 × LK	−0.0022 ***	−0.0010 *	−0.0006	−0.0004
	(−2.64)	(−1.91)	(−1.49)	(−1.62)
Event8 dummy	−0.0025 ***	−0.0018 ***	−0.0005	−0.0014 ***
	(−4.16)	(−3.90)	(−0.98)	(−8.96)
Event8 × LK	0.0088 ***	0.0049 ***	0.0006	0.0042 ***
	(10.24)	(7.90)	(1.14)	(17.06)
Fear Index	0.0001 ***	0.0001 ***	0.0000 ***	0.0000 ***
	(40.80)	(30.45)	(16.36)	(9.69)
Price	−0.0029 ***	−0.0006 ***	−0.0010 ***	0.0004 **
	(−23.97)	(−7.25)	(−9.87)	(2.51)
Volatility	11.9506 ***	10.6716 ***	5.6528 ***	4.9266 ***
	(4.84)	(4.60)	(4.15)	(3.32)
Log(volume)	−0.0051 ***	−0.0036 ***	−0.0026 ***	−0.0010 ***
	(−97.23)	(−78.97)	(−59.94)	(−16.33)
Constant	0.0845 ***	0.0597 ***	0.0418 ***	0.0176 ***
	(102.19)	(82.02)	(61.15)	(17.36)
Observations	117,838	117,709	117,697	117,696
Adjusted R2	0.4037	0.3806	0.1802	0.0279

Table 4

Regression results on spillover effect. This table reports the spillover effect of the Luckin scandal on other Chinese Listings in the U.S. Standard errors are adjusted for heteroscedasticity (Huber–White estimators). *** and ** indicate that the coefficients are statistically significant at 1% and 5% levels, respectively.

	(1)	(2)	(3)	(4)	(5)	(6)
Dependent Variables	(Quoted Spread)	(Quoted Spread)	(Quoted Spread)	(Effective Spread)	(Effective Spread)	(Effective Spread)
Event dummy	−0.0005 **		−0.0003	−0.0004 ***		−0.0003 ***
	(−2.14)		(−1.33)	(−4.65)		(−3.90)
China dummy		0.0045 ***	0.0046 ***		0.0027 ***	0.0027 ***
		(19.80)	(19.30)		(39.80)	(38.78)
Event × China			−0.0012			−0.0004
			(−1.60)			(−1.45)
Fear Index	0.0001 ***	0.0001 ***	0.0001 ***	0.0000 ***	0.0000 ***	0.0000 ***
	(41.57)	(41.75)	(41.55)	(52.28)	(52.96)	(52.39)
Price	−0.0029 ***	−0.0029 ***	−0.0029 ***	0.0007 ***	0.0007 ***	0.0007 ***
	(−23.93)	(−23.54)	(−23.55)	(15.80)	(16.66)	(16.64)
Volatility	11.9564 ***	11.8779 ***	11.8757 ***	2.0221 ***	1.9750 ***	1.9737 ***
	(4.84)	(4.84)	(4.84)	(4.29)	(4.29)	(4.29)
Log(volume)	−0.0051 ***	−0.0050 ***	−0.0050 ***	−0.0025 ***	−0.0024 ***	−0.0024 ***
	(−97.22)	(−96.85)	(−96.85)	(−212.01)	(−210.71)	(−210.73)
Constant	0.0847 ***	0.0828 ***	0.0828 ***	0.0422 ***	0.0411 ***	0.0411 ***
	(102.15)	(101.37)	(101.32)	(219.43)	(215.89)	(215.82)
Observations	117,745	117,745	117,745	117,745	117,745	117,745
Adjusted2	0.4034	0.4068	0.4068	0.5976	0.6054	0.6055
	(1)	(2)	(3)	(4)	(5)	(6)
Dependent Variables	(Realized Spread)	(Realized Spread)	(Realized Spread)	(Price Impact)	(Price Impact)	(Price Impact)
Event dummy	−0.0002		−0.0001	−0.0001 ***		−0.0001 ***
	(−1.04)		(−0.37)	(−3.31)		(−3.33)
China dummy		0.0026 ***	0.0027 ***		0.0009 ***	0.0009 ***
		(12.81)	(12.44)		(26.98)	(26.07)
Event × China			−0.0009			0.0000
			(−1.45)			(0.02)
Fear Index	0.0000 ***	0.0000 ***	0.0000 ***	0.0000 ***	0.0000 ***	0.0000 ***
	(16.92)	(16.85)	(16.84)	(45.84)	(46.20)	(45.76)
Price	−0.0010 ***	−0.0010 ***	−0.0010 ***	0.0007 ***	0.0007 ***	0.0007 ***
	(−9.85)	(−9.56)	(−9.56)	(27.43)	(28.02)	(28.01)
Volatility	5.6542 ***	5.6044 ***	5.6032 ***	0.3944 ***	0.3784 ***	0.3780 ***
	(4.15)	(4.14)	(4.14)	(3.59)	(3.52)	(3.52)
Log(volume)	−0.0026 ***	−0.0025 ***	−0.0025 ***	−0.0006 ***	−0.0006 ***	−0.0006 ***
	(−59.95)	(−59.34)	(−59.35)	(−120.98)	(−118.19)	(−118.18)
Constant	0.0418 ***	0.0407 ***	0.0407 ***	0.0111 ***	0.0107 ***	0.0107 ***
	(61.04)	(60.06)	(60.04)	(130.87)	(126.59)	(126.54)
Observations	117,604	117,604	117,604	117,745	117,745	117,745
Adjusted2	0.1801	0.1823	0.1823	0.3082	0.3146	0.3147

Table 5

Robustness test: regression results on Satyam. This table reports the robust evidence of the regression results for another event of Satyam. Standard errors are adjusted for heteroscedasticity (Huber–White estimators). ***, **, and * indicate that the coefficients are statistically significant at 1%, 5%, and 10% levels, respectively.

	(1)	(2)	(3)	(4)
Dependent Variables	(Quoted Spread)	(Effective Spread)	(Realized Spread)	(Price Impact)
Event dummy	−0.0003	−0.0002	−0.0002	0.0000
	(−0.64)	(−0.55)	(−0.41)	(0.05)
Price	−0.0099 **	−0.0022	−0.0003	−0.0019
	(−2.52)	(−0.76)	(−0.07)	(−0.67)
Volatility	279.5512 *	245.3033	−105.6387	351.0446 **
	(1.68)	(1.62)	(−1.38)	(2.51)
Log(volume)	−0.0014 ***	−0.0009 ***	−0.0006 ***	−0.0003 *
	(−5.82)	(−4.63)	(−2.90)	(−1.93)
Constant	0.0262 ***	0.0169 ***	0.0105 ***	0.0064 **
	(6.08)	(4.81)	(2.83)	(2.28)
Observations	66	66	66	66
Adjusted R2	0.7022	0.7146	0.1887	0.4715

Appendix A. Internet Appendix

Table A1

Variable descriptions.

Variables	Definition
China Dummy	A dummy variable if the enlisted stock is headquartered in China
Event Dummy	Dummy variables for eight events.
Fear Index	The fear index is the equally weighted measure of the reported case index (RCI) and reported death index (RDI), used by Salisu and Akanni (2020), Salisu et al. (2020), and Mazumder and Saha (2021)
Log (Volume)	Logarithm of the average daily dollar trading volume
Price	The share price of a given asset
Volatility	Standard deviation of intraday quote-midpoint returns
Quoted Spread	$\left(As{k}_{j,t}-Bi{d}_{j,t}\right)/Mi{d}_{j,t}$
Effective Spread	$2D_{j,t}\left(P_{j,t}-Mi{d}_{j,t}\right)/Mi{d}_{j,t}$
Realized Spread	$2D_{j,t}\left(P_{j,t}-Mi{d}_{j,t+5}\right)/Mi{d}_{j,t}$
Price Impact	$100D_{j,t}\left(M_{j,t+5}-Mi{d}_{j,t}\right)/Mi{d}_{j,t}$

Table A2

Event dates and descriptions.

Dates	Description
8 January 2020	Luckin announced its pricing of a follow-on offering of ADRs
14 January 2020	Luckin announced its completion of a follow-on offering of ADRs
31 January 2020	Muddy Waters Research released an 89-page report and declared it planned to short sell Luckin
4 February 2020	Ash Illumination Research released a 49-page report in Chinese, highlighting Luckin’s financial fraud and released a 66-page report in English
2 April 2020	Luckin admitted that it overstated its profits by RMB 2.2 billion (USD 310 million)
20 May 2020	Luckin resumed trading
23 june 2020	Luckin received the second delisting notice from Nasdaq
26 June2020	Luckin announced a trading half on June 29th and would delist soon

References

Ahmad, Syed R.; Olarewaju, Odunayo M.; Ali, Ijaz; Baig, Asif; Khan, Imran A. Impact of accounting fraud on stock price formation before its discovery- the period from the start of fraud to its discovery. Academy of Entrepreneurship Journal; 2021; 27, pp. 1-18.

Akhigbe, Aigbe; Madura, Jeff; Martin, Anna. Accounting contagion: The case of Enron. Journal of Economics and Finance; 2005; 29, pp. 187-202. [DOI: https://dx.doi.org/10.1007/BF02761553]

Beatty, Anne; Liao, Scott; Yu, Jeff J. The spillover effect of fraudulent financial reporting on peer firms’ investments. Journal of Accounting and Economics; 2013; 55, pp. 183-205. [DOI: https://dx.doi.org/10.1016/j.jacceco.2013.01.003]

Beneish, Messod D.; Lee, Charles; Nichols, David C. Fraud Detection and Expected Returns. SSRN 1998387. 2012; Available online: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=1998387 (accessed on 10 January 2023).

Brown, Stephen V.; Tian, Xiaoli; Tucker, Jennifer W. The spillover effect of SEC comment letters on qualitative corporate disclosure: Evidence from the risk factor disclosure. Contemporary Accounting Research; 2018; 35, pp. 622-56. [DOI: https://dx.doi.org/10.1111/1911-3846.12414]

Chen, Tianhao. Blockchain and accounting fraud prevention: A case study on Luckin Coffee. Paper presented at the 2022 7th International Conference on Social Sciences and Economic Development (ICSSED 2022); Wuhan, China, March 25–27; Atlantis Press: Amsterdam, 2022; pp. 44-49.

Clayman, Michelle R.; Fridson, Martin S.; Troughton, George H. Corporate Finance: A Practical Approach; John Wiley & Sons: Hoboken, 2012.

Darrough, Masako; Huang, Rong; Zhao, Sha. Spillover effects of fraud allegations and investor sentiment. Contemporary Accounting Research; 2020; 37, pp. 982-1014. [DOI: https://dx.doi.org/10.1111/1911-3846.12541]

Donley, Lacy; Legoria, Joseph; Reichelt, Kenneth J.; Walton, Stephanie. Chinese auditor inspection access challenges: The market’s response to integrated US regulatory and legislative action. Journal of Accounting and Public Policy; 2023; 42, 107110. [DOI: https://dx.doi.org/10.1016/j.jaccpubpol.2023.107110]

Dumontaux, Nicholas; Pop, Adrian. Understanding the market reaction to shockwaves: Evidence from the failure of Lehman Brothers. Journal of Financial Stability; 2013; 9, pp. 269-86. [DOI: https://dx.doi.org/10.1016/j.jfs.2013.04.001]

Ellis, Katrina; Michaely, Roni; O’Hara, Maureen. The accuracy of trade classification rules: Evidence from NASDAQ. Journal of Financial and Quantitative Analysis; 2000; 35, pp. 529-52. [DOI: https://dx.doi.org/10.2307/2676254]

Griffin, Paul A.; Grundfest, Joseph A.; Perino, Michael A. Stock price response to news of securities fraud litigation: An analysis of sequential and conditional information. Abacus; 2004; 40, pp. 21-48. [DOI: https://dx.doi.org/10.1111/j.1467-6281.2004.00149.x]

Kim, Jang-Chul; Su, Qing; Elliott, Teressa. The impact of democracy on liquidity and information asymmetry for NYSE cross-listed stocks. International Review of Finance; 2024a; [DOI: https://dx.doi.org/10.1111/irfi.12469]

Kim, Jang-Chul; Mazumder, Sharif; Su, Qing. Brexit’s ripple: Probing the impact on stock market liquidity. Finance Research Letters; 2024b; 61, 105030. [DOI: https://dx.doi.org/10.1016/j.frl.2024.105030]

Lamoreaux, Phillip T. Does PCAOB inspection access improve audit quality? An examination of foreign firms listed in the United States. Journal of Accounting and Economics; 2016; 61, pp. 313-37. [DOI: https://dx.doi.org/10.1016/j.jacceco.2016.02.001]

Leung, Danny. Expresso-Charged IPO: Luckin Coffee to Set a New World Record. FinanceAsia. 2019; Available online: https://www.financeasia.com/article/espresso-charged-ipo-luckin-coffee-to-set-a-new-world-record/451530 (accessed on 4 November 2024).

Li, Shaomin; Wu, Judy J. Corruption: Why China thrives despite corruption. Far East Economic Review; 2007; 170, pp. 1-6.

Lim, Lionel. Luckin Coffee, the Buzzy Chain That Outsells Starbucks in China, Reportedly Plans a U.S. Expansion. Fortune.com. 2024; Available online: https://fortune.com/asia/2024/10/29/luckin-coffee-us-expansion-starbucks-china-competition-cotti/ (accessed on 4 November 2024).

Mazumder, Sharif; Saha, Pritam. COVID-19: Fear of pandemic and short-term IPO performance. Finance Research Letters; 2021; 43, 101977. [DOI: https://dx.doi.org/10.1016/j.frl.2021.101977]

Morris, Brandon C.; Egginton, Jared F.; Fuller, Kathleen P. Return and liquidity response to fraud and SEC investigations. Journal of Economics and Finance; 2019; 43, pp. 313-29. [DOI: https://dx.doi.org/10.1007/s12197-018-9445-y]

Peng, Zhe; Yang, Yahui; Wu, Renshui. The Luckin Coffee scandal and short-selling attacks. Journal of Behavioral and Experimental Finance; 2022; 34, 100629. [DOI: https://dx.doi.org/10.1016/j.jbef.2022.100629]

Public Company Accounting Oversight Board (PCAOB). PCAOB Release No. 2021-001. 2021; Available online: https://assets.pcaobus.org/pcaob-dev/docs/default-source/rulemaking/docket048/2021-001-hfcaa-proposing-release.pdf?sfvrsn=dad8edcf_6 (accessed on 10 March 2024).

Richardson, Grant; Obaydin, Ivan; Liu, Chelsea. The effect of accounting fraud on future stock price crash risk. Economic Modelling; 2022; 117, 106072. [DOI: https://dx.doi.org/10.1016/j.econmod.2022.106072]

Sadka, Gil. The economic consequences of accounting fraud in product markets: Theory and a case from the U.S. telecommunications industry (WorldCom). American Law and Economics Review; 2006; 8, pp. 439-75. [DOI: https://dx.doi.org/10.1093/aler/ahl012]

Salisu, Afees A.; Akanni, Lateef O. Constructing a global fear index for the COVID-19 pandemic. Emerging Markets Finance Trade; 2020; 56, pp. 2310-31. [DOI: https://dx.doi.org/10.1080/1540496X.2020.1785424]

Salisu, Afees A.; Akanni, Lateef; Raheem, Ibrahim. The COVID-19 global fear index and the predictability of commodity price returns. Journal of Behavioral and Experimental Finance; 2020; 27, 100383. [DOI: https://dx.doi.org/10.1016/j.jbef.2020.100383]

Wang, Qiyao. Cost of the accounting scandal of Luckin Coffee to multiple aspects and the influence under current economy and pandemic time. Paper presented at the 2020 2nd International Conference on Economic Management and Cultural Industry (ICEMCI 2020); Chengdu, China, October 23–25; Atlantis Press: Amsterdam, 2020; pp. 170-73.

Weske, Jennifer; Benuto, Lorainne. Share prices and price/earnings ratios as predictors of fraud prior to a fraud announcement. Academy of Accounting and Financial Studies Journal; 2015; 19, pp. 281-97.

Wu, Joanna. Villains or Victims?: What to Make of U.S.-Listed Chinese Companies. 2022; Available online: https://simon.rochester.edu/blog/deans-corner/villains-or-victims-what-make-us-listed-chinese-companies (accessed on 15 February 2023).

Yang, Dan; Jiao, Hao; Buckland, Roger. The determinants of financial fraud in Chinese firms: Does corporate governance as an institutional innovation matter?. Technological Forecasting and Social Change; 2017; 125, pp. 309-20. [DOI: https://dx.doi.org/10.1016/j.techfore.2017.06.035]

Zhang, Yiqian; Obot, Iberedem. Luckin Coffee: A look at corporate governance in the Chinese market. Cases on Uncovering Corporate Governance Challenges in Asian Markets; IGI Global: Pennsylvania, 2024; pp. 77-93.

Zhu, Julie. Luckin Coffee’s Journey from Hit Startup to $5 Billion Share Wipeout. Reuters. 2020; Available online: https://www.reuters.com/article/business/luckin-coffees-journey-from-hot-startup-to-5billion-share-wipeout-idUSKBN21L1HW/ (accessed on 4 November 2024).