Knowledge Capital and Stock Returns during Crises in the Manufacturing Sector: Moderating Role of Market Share, Tobin’s Q, and Cash Holdings

Lee, Chaeho Chase; Atukeren, Erdal; Kim, Hohyun

doi:10.3390/risks12060100

Open AccessArticle

Knowledge Capital and Stock Returns during Crises in the Manufacturing Sector: Moderating Role of Market Share, Tobin’s Q, and Cash Holdings

by

Chaeho Chase Lee

^1,2,

Erdal Atukeren

¹

and

Hohyun Kim

^3,*

¹

BSL Business School Lausanne, Route de la Maladière 21, 1022 Chavannes, Switzerland

²

Seoul Business School, aSSIST University, 46, Ewhayeodae 2-gil, Seodaemun-gu, Seoul 03767, Republic of Korea

³

School of Management and Economics, Handong Global University, 558, Handong-ro, Heunghae-eup, Buk-gu, Pohang-si 37554, Gyeongsangbuk-do, Republic of Korea

^*

Author to whom correspondence should be addressed.

Risks 2024, 12(6), 100; https://doi.org/10.3390/risks12060100

Submission received: 25 April 2024 / Revised: 29 May 2024 / Accepted: 14 June 2024 / Published: 19 June 2024

Download Versions Notes

Abstract

:

This study analyzes the impact of knowledge capital (KC), a key element of firms’ innovation and competitiveness, on stock returns during economic crises when sustainable competitiveness becomes particularly important. We analyze the impact of the Global Financial Crisis and COVID-19 as economic crises, focusing on manufacturing industries with a high proportion of investment shifts from physical capital to KC. Our findings indicate that KC is positively associated with stock returns during the Global Financial Crisis and COVID-19. This positive relationship is strengthened by the firm’s ability to leverage KC, as measured by greater product market share, higher Tobin’s Q, and larger cash holdings. This study emphasizes the protective role of KC during the economic crisis when the market pays more attention to corporate sustainability and provides implications to corporate managers and investors.

Keywords:

knowledge capital; economic crisis return; market share; Tobin’s Q; cash holdings

1. Introduction

Economic crises offer a distinct opportunity to examine a firm’s sustainable competitiveness (Bose et al. 2022; Lins et al. 2017; Lome et al. 2016). Such crises, characterized by credit risks, economic lockdowns, and fluctuations in global demand and trade, significantly challenge corporate sustainability (Bekaert et al. 2014; Calomiris et al. 2012; Ding et al. 2021; Fahlenbrach et al. 2021). Researchers frequently use these experimental settings to investigate what drives corporate sustainability, especially as investors focus more on sustainable firms during these turbulent times (Hartzmark and Sussman 2019; Hasan and Uddin 2022). This study leverages recent crises, specifically the Global Financial Crisis (GFC) and the COVID-19 pandemic, to explore the relationship between corporate sustainability and a less-examined intangible capital—knowledge capital (KC).

KC represents a corporate capacity to innovate products and processes, which is fundamental for sustaining a long-term competitive advantage. Empirically, KC is estimated by accumulated historical R&D expenditures (Peters and Taylor 2017; Sanford and Yang 2022). Companies with more KC have higher corporate value and better operational performance in the long term (Chan et al. 2001; Chiu et al. 2021). In industries with intense competition, companies with abundant KC have shown higher competitiveness (Gu 2016; Kim et al. 2021). From a manufacturing perspective, companies can enhance production efficiency by leveraging KC, thereby maintaining higher KC long-term and accruing greater profits (Peters et al. 2017; Ugur and Vivarelli 2021). Therefore, such companies possess higher value and better production efficiency. Additionally, in an analysis focusing on the U.S. manufacturing sector, firms with higher KC mitigate performance declines caused by import competition from emerging markets (Hombert and Matray 2018; Sheng and Montgomery 2022). They also achieve higher productivity, sales growth, and profitability based on innovation capabilities derived from KC.

Existing studies have identified the impact of KC on stock returns during non-crisis periods based on sustainable competitiveness. However, a research gap remains in understanding how KC fundamentally influences a firm’s sustainable competitiveness, particularly through stock returns during economic downturns. During periods of crisis, unexpected declines in overall market confidence make shareholders more concerned that the financial information they previously relied on for investment decisions may no longer be trustworthy or valid (Lins et al. 2017; Trinh et al. 2023). As a result, the extent to which financial information is reflected in stock prices becomes limited (Hartzmark and Sussman 2019; Yoo et al. 2021). Therefore, the period of economic crisis is crucial for investigating the role of KC as a determinant of a firm’s sustainability, as stock prices during this time highlight sustainable competitive advantages more prominently. However, research on the impact of KC on stock returns during economic crises is lacking, and particularly, analyzing the manufacturing industry, where productivity enhancements critically affect competitiveness, is a first in this study.

Our empirical analysis focuses on the impact of KC on crisis-period returns within the US manufacturing sector. We choose this sector due to its recent shift from investments in physical capital (PC) to those in KC (Antonelli 2019; Corrado et al. 2022). Previous studies on US-listed firms have shown a notable increase in the output elasticity of KC and a corresponding decrease in PC, especially pronounced in the manufacturing sector (Antonelli et al. 2023). Unlike PC, which can be easily traded in the open market, KC involves unique firm-specific attributes and thus cannot be easily bought or sold (Dierickx and Cool 1989; Pereira and Bamel 2021). While emerging economies strategically build manufacturing capabilities through PC investments, developed countries have increasingly outsourced PC-centric activities within the global value chain and shifted their core operations toward more knowledge-intensive activities like R&D (Boehm et al. 2020). This nuanced shift underscores the critical role of KC in fostering sustainability and resilience in the manufacturing sector during economic crises.

Even for firms with the same level of KC, the performance of KC may vary depending on the firm’s market leadership, value-creation capabilities, and financial resources. Therefore, we investigate the potential moderating factors in KC utilization and explore their impact on stock returns during periods of crisis: market share, Tobin’s Q, and cash holdings. Market share indicates a firm’s market leadership. It is positively correlated with the success of new products and services, suggesting that firms with higher market shares can more effectively convert KC into profits (Bhattacharya et al. 2022; Hula 1989; Willekens et al. 2023). Tobin’s Q, an indicator of a firm’s growth opportunities, suggests that firms with higher values are recognized for generating more value than the investments made, including those in KC (Coluccia et al. 2020; Parcharidis and Varsakelis 2010). We also consider the role of cash holdings, which can reduce financial constraints, supporting sustained investment in KC and ongoing R&D efforts (Chang and Yang 2022; He and Wintoki 2016).

Our empirical findings reveal a robust positive relationship between KC and stock returns during the two crises. These findings suggest that investors prefer firms with substantial KC during economic downturns, viewing it as crucial for corporate sustainability. Economically, a one-standard-deviation increase in KC was associated with 7.11 and 3.24 percentage points in crisis returns during the GFC and COVID-19 periods, respectively. Additionally, the analysis shows that higher market share, Tobin’s Q, and cash holdings amplify the positive impact of KC on crisis returns. In contrast, firms with lower values in these areas do not exhibit significant effects of KC. Robustness tests with Two-Stage Least Squares (2SLS) regressions support the reliability of our findings.

These findings contribute to the existing literature on KC and firm values. Previous studies have primarily focused on the role of KC in increasing market value during non-crisis periods (Chiu et al. 2021; Laperche 2021). Our study extends this understanding by examining the relationship during crisis periods, elucidating KC’s role in corporate sustainability. Moreover, our research provides insights into leveraging KC by effectively managing market share, Tobin’s Q, and cash holdings, suggesting that strategic KC management can maximize corporate sustainability. Our study advises investors to consider KC a crucial factor in managing stock portfolios during crises, as it appears to act as a buffer against external economic shocks. Allocating more resources to firms with robust KC could yield better performance in turbulent times.

The subsequent sections of the paper are structured as follows: Section 2 provides an overview of the empirical methodology, encompassing data sources and variables. Section 3 discusses our results of the impact of KC on crisis returns and details the robustness tests conducted. Finally, in Section 4, we draw our conclusions.

2. Data and Methods

2.1. Sample and Data

This study focuses on two significant economic crises: the GFC and the COVID-19 pandemic. We define the GFC period from 15 September 2008, marking Lehman Brothers’ bankruptcy, to 9 March 2009, when the S&P 500 and MSCI World Equity Index reached their lowest points (Frankel and Saravelos 2012). For the COVID-19 period, we concentrate on the dates from 19 February to 23 March 2020, which represent the peak and trough of the S&P 500 index during the early phase of the pandemic (Cox et al. 2020). We analyze crisis and non-crisis periods to examine how KC impacts crisis returns at the firm level. For each crisis, the analysis includes a non-crisis period: for the GFC, we include the four years before and after the crisis, and for COVID-19, we include the two years before and after the crisis. The sample for the COVID-19 period ranges from 2018 to 2022, restricted to two years on either side of the crisis due to data availability. We specifically analyze firms within the SIC two-digit codes 31 to 39 to understand the relationship between KC and crisis returns based on the characteristics of the manufacturing industry.

We gathered daily stock return data from the CRSP database and annual accounting figures from COMPUSTAT based on the fiscal year ending before each testing year. To ensure the reliability and consistency of our study during the GFC and COVID-19 periods, we excluded samples that could distort the analysis result (Sha et al. 2022; Zhang et al. 2024). We excluded small-cap stocks—those with a market capitalization below USD 250 million at the end of the preceding year—for both the GFC and COVID-19 periods because of their limited liquidity, broader bid-ask spreads, and increased vulnerability to trading-induced price pressures. The final sample consists of 1105 firms for the GFC period and 824 firms for the COVID-19, respectively.

2.2. Variables

2.2.1. Dependent Variable; Stock Returns

This study employs two stock return variables: raw return and abnormal return. Raw return is based on daily excess returns during the GFC and COVID-19 periods. These excess returns are calculated by subtracting the risk-free rate, represented by the 1-month daily Treasury bill rate, from the daily buy-and-hold returns. For the GFC, these returns are calculated from 15 September of the test year to 9 March of the following year. For the COVID-19 crisis, stock returns are estimated from 19 February to 23 March 2020, along with returns from years immediately before and after this period. Abnormal return measures the difference between the firm’s expected and raw returns. The expected return is calculated using the market model with the CRSP value-weighted index as the market proxy. We estimate the expected return over a 60-month period, ending the day before the start of the raw return period for each respective year.

2.2.2. Main Independent Variable; Knowledge Capital

We measure KC with the perpetual inventory method by following the approach in previous literature (Eisfeldt and Papanikolaou 2013; Peters and Taylor 2017). This method involves the aggregation of deflated R&D expenses, as shown in Equation (1):

{K C}_{i, t} = {K C}_{i, t - 1} (1 - δ_{0}) + \frac{R & D_{I, t}}{C P I_{t}}

(1)

where i denotes the firm, t represents the fiscal year, δ₀ (set at 15%) is the rate reflecting the estimation of the depreciation used by the US Bureau of Economic Analysis for R&D capital (Ewens et al. 2024; Peters and Taylor 2017), and CPI is the consumer price index, which adjusts for inflation. For cases where R&D expense data are missing in COMPUSTAT, we employ a method that Peters and Taylor (2017) suggested. This method estimates missing R&D values using age-specific growth rates derived from SIC two-digit industrial classifications. Additionally, estimating the initial stock of KC is necessary when accounting data from a firm’s founding year are unavailable. This estimation follows the prior approach (Eisfeldt and Papanikolaou 2013; Peters and Taylor 2017), as laid out in Equation (2):

{KC}_{i, t 0} = \frac{R & D_{i, t 1}}{g + δ_{0}}

(2)

where t₁ is the first year of a firm i appears in our sample, and g represents the estimated average growth rate of R&D expenses before its appearance in the dataset, calculated as approximately 10.08%. We normalize each firm’s KC by its total assets to facilitate comparisons across firms. This normalization ensures that the KC values are comparable regardless of the firm’s size.

2.2.3. Moderators; Market Share, Tobin’s Q, and Cash Holdings

Market share is defined as the annual revenue of each firm expressed as a proportion of the total revenue within their respective industries, based on SIC two-digit levels. We classify companies with a market share above the industry-specific median yearly as high market share firms. Conversely, those with a market share below this median are considered low-market share firms. Tobin’s Q is a measure of firm valuation calculated by dividing the sum of the market value of equity and total debt by total assets (Coluccia et al. 2020; Wright 2004). To categorize firms based on Tobin’s Q, we use their SIC two-digit classification to assign them into either high or low groups, depending on whether Tobin’s Q is above or below the median value within their industry. Cash holdings are measured by the ratio of cash and cash equivalents to total assets. Firms are categorized as having high or low cash holdings based on whether this ratio is above or below the median within their respective SIC two-digit industry classification.

2.3. Empirical Model

Our empirical analysis employs an ordinary least squares (OLS) regression, as illustrated in Equation (3):

Return_i,t = β₀ + β₁·KC_i,t−1 + β₂ Crisis _t + β₃ KC_i,t−1 × Crisis_t + β₄ X_i,t−1 + β₅ Factor Loadings_i,t + Firm Fixed Effects + e_i,t

(3)

where Return_i,t denotes the daily stock return, KC_i,t−1 represents the firm’s KC, measured at the last fiscal year before the return measurement, and Crisist is a binary variable that equals one during the crisis periods and zero otherwise. The β3 coefficient captures the interaction of KC and crisis, indicating the impact of KC on firm value during economic crises.

We control various accounting- and market-level variables. Market capitalization is the logarithm of the product of the closing price and total number of outstanding shares. Long-Term Debt is calculated as the ratio of long-term debt to total assets. Short-Term Debt is the ratio of current liabilities related to debt to total assets. Cash holdings are measured by the ratio of cash and cash equivalents to total assets. Profitability is the ratio of operating income to total assets. Companies with high profitability, abundant cash holdings, and low debt can continue investing during economic crises. At the same time, other firms may reduce investments, especially if they face short-term debt maturities during the crisis. Therefore, we employ these as control variables (Afiezan et al. 2020; Harford et al. 2014).

Additionally, we employ characteristics that may influence firm-specific stock returns as control variables—the book-to-market ratio roles as an indicator of whether a company is overvalued or undervalued. Previous studies have identified a significant positive relationship between the book-to-market ratio and stock returns (Daniel and Titman 1997; Nugroho 2020). Therefore, this study uses the book-to-market ratio as a control variable to examine the impact of KC on stock returns. We also add a dummy variable for firms with a negative book-to-market ratio. Firms with a negative book-to-market ratio are likely to be in financial distress, and their stock returns may develop differently than those with a positive book-to-market ratio (Fama and French 1992; Jan and Ou 2020). Such firms are likely to be financially disadvantaged. As a result, their returns may exhibit characteristics similar to firms with a high book-to-market ratio rather than those with a low book-to-market ratio (Fama and French 1992; Habib et al. 2022). Under the premise that idiosyncratic risk may also affect stock returns, we control for a firm’s idiosyncratic risk (Li et al. 2021; Lins et al. 2017).

The book-to-market ratio is computed as the ratio of equity’s book value to its market value. The Negative B/M is a binary indicator that takes one when the book-to-market ratio is negative and zero otherwise. Idiosyncratic risk is measured as the variance of residuals from the market model’s expected returns over sixty months leading up to the day before the analysis period. Furthermore, we include factor loadings from the Fama–French three-factor model plus a momentum factor (Eisfeldt et al. 2020; Fama and French 1992), recalibrated annually based on data from the previous 60 months. Factor loadings from these four factors are incorporated in regression models linearly. We incorporate industry dummy variables at the two-digit SIC level to account for variations in KC investments across sectors and their potential differential impacts during crises.

To mitigate the issue of autocorrelation in panel data analysis, where the error term at one point in time may be correlated with the error term from a previous point, leading to an underestimation of the standard errors in OLS, we cluster the standard errors at the firm level in all our regression analyses. In some models, industry- or firm-fixed effects control for unobserved, industry- or firm-specific influences on the relationship between KC and crisis returns. All variables are winsorized at the 1st and 99th percentiles to mitigate the effect of outliers. Detailed definitions of these variables are provided in Appendix A.

2.4. Summary Statistics

Panels A and B in Table 1 present the summary statistics for variables during the GFC and COVID-19 periods. KC, a measure of KC divided by total assets, has a mean of 0.232 during the GFC and 0.148 during the COVID-19 pandemic. Crisis stock return averages −0.557 and −0.404 during the GFC and COVID-19. The evidence of drastic meltdowns in the stock markets makes it clear why these periods are critical for analyzing the impact of KC on corporate sustainability.

Investment in R&D, a key component of KC, tends to be pro-cyclical (Ahmad and Zheng 2023; Roper and Turner 2020). Therefore, it is necessary to ascertain whether our empirical analysis results could be affected or distorted by differences in KC between economic crisis and non-crisis periods. We present descriptive statistics in Appendix B, derived by composing sub-samples for crisis and non-crisis periods. Upon comparing the KC during economic crisis periods with non-crisis periods, we found that similar levels of KC were observed during both periods.

Panels C and D reveal the correlation matrices for the two crisis periods. KC shows a significant positive correlation with general and crisis-period stock returns, at least at the 5% significance level. This correlation suggests that KC benefits a firm’s valuation in general and, more importantly, corporate sustainability during crises. Other variables, such as firm size, measured by market capitalization, cash holdings, and profitability, also positively correlated with stock returns during these crises. On the other hand, long-term debt and the book-to-market ratio are negatively correlated with stock returns, aligning with typical expectations for crisis periods.

3. Results and Analyses

3.1. KC and Crisis Returns during GFC

We conducted multiple regression analyses to explore the influence of a company’s KC on stock returns during crisis periods. Table 2 presents the results of the regression analysis during the GFC period. Columns (1) through (3) analyze the relationship between KC and raw returns, while columns (4) through (6) analyze the relationship between abnormal returns and KC. Each Raw and abnormal return analysis incorporates industry and firm fixed effects. Hausman test results suggest the adoption of firm fixed effects models for the GFC period (χ² (12) = 65.46, p-value < 0.01) (Hausman 1978).

The relationship between KC and stock returns presented in the first row in Table 2 was found to be insignificant. Previous studies analyzing the IT industry (Chiu et al. 2021) and those estimating KC through the accumulation of patents (Hegde and Mishra 2023) have confirmed a positive relationship between KC and stock returns. We hypothesize that the differences in result could be attributed to variations in the periods reviewed for stock returns, the characteristics of the IT industry versus manufacturing industries, and the nature of KC, which includes unsuccessful R&D investments as well as patents. Additionally, including economic crisis periods, outliers in the dependent variable of stock returns could have distorted the results.

In all columns, the interaction term between KC and crisis returns has positive and significant coefficients, indicating KC’s positive impact on stock returns during crises. KC improves innovation and operational performance in the long term, serving as a key factor in sustainable competitive advantage. Therefore, it acts as a buffer against stock price declines during crises when investors’ confidence in existing performance diminishes. The size of coefficients in columns (3) and (6) illustrates the impact of KC on raw return and abnormal return when applying the firm fixed effects model. During the GFC period, a one standard deviation increase in KC (0.549) is associated with a 7.36 percentage point increase in raw return and a 5.93 percentage point increase in abnormal return. Overall, the results in Table 2 highlight investors’ heightened importance on KC during crisis periods. It supports that robust KC reserves are crucial for firm resilience during financial crises.

3.2. KC and Crisis Returns during COVID-19

We verify the results from the GFC period with another recent economic crisis, COVID-19. Table 3 illustrates the relationship between KC and both raw returns and abnormal returns during COVID-19. Columns (1) to (3) analyze the relationship between KC and raw returns, and columns (4) to (6) examine the same variable with abnormal returns. All results from columns (1) to (6) display a pattern similar to that of the GFC, showing a positive correlation between KC and both raw and abnormal returns during COVID-19. Specifically, columns (3) and (6), which reflect firm fixed effects and control variables, indicate that a one standard deviation increase in KC (0.214) results in a 3.38 percentage point increase in raw returns and a 5.16 percentage point increase in abnormal returns.

These findings suggest that KC acts as a corporate immune system during external economic shocks, providing investors confidence and mitigating corporate value declines. Therefore, firms with higher KC tend to experience relatively smaller declines in stock returns during economic crises. Our assertion that firms with more accumulated KC during crisis periods will have higher stock returns is supported by a 1% statistical significance level.

3.3. Moderating Role of Market Share, Tobin’s Q and Cash Holdings

Table 4 provides the results from a subsample regression analysis designed to explore the moderating effects of market share, Tobin’s Q, and cash holdings on the relationship between KC and crisis-period raw returns. Panel A and B represent the results during the GFC and COVID-19 periods, respectively. Columns (1) and (2) of Panel A and B focus on the influence of market share. The analysis reveals that market share plays a significant role in moderating the impact of KC on stock returns during crisis periods. This result is evident in both the GFC and the COVID-19 pandemic. The findings indicate that companies with a higher market share benefit from a stronger positive relationship between KC and crisis returns. The rationale might be that firms with significant market presence are more capable of utilizing their KC to develop products and services that reinforce their market dominance, thereby ensuring stable profitability even during turbulent times.

As reported in Columns (3) and (4) of Panel A and B, the results for Tobin’s Q indicate a significant interaction. Firms with a higher Tobin’s Q demonstrate a stronger linkage between KC and crisis returns. This finding suggests that companies with higher Tobin’s Q are more efficient at creating value from their investments in KC, thereby enhancing their competitiveness sustainably. Finally, Columns (5) and (6) focus on cash holdings. The findings show that higher cash holdings fortify the relationship between KC and crisis returns. Companies with more cash reserves can better sustain the reliability and integrity of their KC, leading to increased sustainable competitiveness. Each of these factors—market share, Tobin’s Q, and cash holdings—plays a critical moderating role, reinforcing the positive impact of KC on stock returns during financial crises. This understanding emphasizes the importance of strategic asset management and capital allocation in enhancing a firm’s resilience and performance in economic adversity.

3.4. Robustness

We attempt to address the potential endogeneity that arises from omitting certain firm-level time-variant characteristics, such as profitability, culture, and stability, which may correlate with KC and affect stock returns during economic crises. Thus, there is a potential risk that these omitted variables might influence or distort our findings (Zaefarian et al. 2017). Moreover, simultaneity bias emerges when KC and crisis returns are jointly determined (Ullah et al. 2018). For instance, firms with large KC are typically more profitable and stable (Chiu et al. 2021; Faucheux 2010).

We implement a 2SLS regression analysis to address these concerns, using instrumental variables to mitigate endogeneity. Based on previous studies (Carlin et al. 2012; Hasan 2018), our first instrumental variable is the industry-level average KC based on two-digit SIC codes for each year. Since intangible capital is often similar within industries, a firm’s KC is expected to correlate with its industry-level counterpart. This industry-specific average is presumed not to directly influence a firm’s crisis returns, satisfying the necessary condition for a valid instrumental variable. Additionally, we employ a 2-year lagged KC as a second instrumental variable, recognizing that the benefits of past R&D investments may take time to materialize and thus influence current investment levels (Chang and Kang 2019; Doraszelski and Jaumandreu 2013) This approach acknowledges the cumulative nature of KC.

Furthermore, we use the average value of state-level KC as a third instrumental variable. According to geographical economics theory (Krugman 1997), namely the theory of industrial agglomeration, firms can benefit from interactions with other nearby firms (Huang and Yuan 2021). This theory implies that other proximate firms’ research and development expenses can influence the effect of a firm’s own KC. Therefore, the average R&D investment of firms within the same state, which results from the accumulation of R&D investments, has been recognized to impact a firm’s KC potentially and has been utilized as an instrumental variable in previous research (Chang 2018).

In Table 5, we report the results of 2SLS regressions with instrumental variables, considering firm fixed effects. Panels A and B present the analysis results for the GFC and COVID-19 periods, respectively. Each panel’s column (1) presents the first-stage regression results examining the relationship between KC and three instrumental variables while controlling for firm characteristics and fixed effects. We observe a significant and positive relationship between KC and the instrumental variables. The model analyzed for the GFC period passes Anderson’s under-identification test (χ² = 289.90, p < 0.01), indicating it is not under-identified. Furthermore, the Cragg–Donald Wald F-statistic is 90.7, exceeding the Stock–Yogo critical value of 22.30, allowing us to reject the null hypothesis of weak instruments (Cragg and Donald 1993; Stock and Yogo 2005). During COVID-19, the analysis also passed the under-identification test (χ² = 696.64, p < 0.01), and the Cragg–Donald Wald F-statistic was 74.1, exceeding the Stock–Yogo critical value of 22.30.

Each panel’s columns (2) and (3) present the results of the second-stage regressions of raw return and abnormal return, using the fitted value of KC for the crisis return measures. Consistent with the results in Table 2 and Table 3, the instrumented KC exhibits a positive and statistically significant relationship with both raw and abnormal returns during the crisis periods. In the second-stage regressions analyzed for the GFC and COVID-19 periods, the p-values of Hansen’s J over-identification test statistic are large: 0.42 for raw return and 0.34 for abnormal return during the GFC period, and 0.34 and 0.37, respectively, during the COVID-19 period. These results indicate that the instrumental variables are significantly uncorrelated with the error term, suggesting their validity (Hansen 1982). The results of the 2SLS analysis are consistent with those presented in Table 2 and Table 3, affirming that KC exhibits a positive and statistically significant relationship with crisis-period stock returns.

Furthermore, to enhance our approach to addressing endogeneity, we utilized the Lewbel (2012) method, which accounts for heteroskedasticity within the regression model. Instead of estimating endogenous variables based on instrumental and exogenous variables in the first stage, this method modifies the regression model by considering heteroskedasticity assumptions within the model. This method has been applied in recent research within the corporate finance field (Chen et al. 2021; Hasan et al. 2021; Mavis et al. 2020). Columns (3) and (6) present the results. We find that instrumented KC using Lewbel’s estimation method exhibits consistently positive and statistically significant relationships with raw and abnormal returns during the GFC and COVID-19 crises.

The linear models based on OLS are susceptible to distortion when there are many explanatory variables (Zhang et al. 2019). Particularly when the signal-to-noise ratio is very low, these models overfit noise rather than extracting the signal. To address the overfitting issue, we utilized Ridge and Lasso regression analyses. These models improve the stability of out-of-sample predictions by shrinking the penalized linear regression coefficients, thereby reducing the number of estimated parameters (Altelbany 2021; McNeish 2015). Upon reviewing Ridge and Lasso regression to assess the potential for overfitting, our study found that the explanatory power remained comparable, confirming a low likelihood of overfitting.

In empirical studies based on panel data, heteroscedasticity–robust standard errors are often reported (MacKinnon et al. 2023; White 1980). An alternative commonly used in research is reporting cluster–robust standard errors, where clustering is often applied at geographic units, such as the state or county level and industry level (Abadie et al. 2023; Arellano 1987). We conducted robustness tests in our study by clustering the results at a broader level, at the industry and state level, to ensure the robustness of the results. Analyzing the data with industry and state-level clustering yielded statistically significant results and confirmed the consistency with the original findings. The test results for cluster–robust standard errors are presented in Appendix C.

4. Conclusions

This study analyzed the impact of KC on stock returns during economic crises, aiming to confirm whether KC is a core factor influencing a firm’s long-term sustainable competitiveness through facilitating innovation in products and processes within firms. The economic crises during the periods of the GFC and COVID-19 served as opportune moments to analyze the impact of KC on a company’s sustainable competitiveness, as investors prioritize sustainable competitiveness over disclosed financial performance when evaluating a company’s market value. Our analysis supports the hypothesis that KC benefits stock returns in times of economic crisis. Specifically, we observe that a one-standard-deviation increase in KC is associated with 6.02 percentage points in stock returns during the GFC and a 3.38 percentage point increase during the COVID-19 crisis. These findings highlight the protective role of KC as a buffer against external economic shocks.

Furthermore, we analyzed the moderating effects of market share, Tobin’s Q, and cash holdings to explore the conditions under which KC better serves as a buffer in stock returns during crisis periods. The study found that firms with higher market share, Tobin’s Q, and cash holdings experience a stronger positive association between KC and stock returns during crises. This finding indicates that a firm’s market leadership, value creation capabilities, and financial resources are facilitating factors for KC to generate sustainable competitiveness.

Our empirical findings significantly contribute to the literature by quantitatively establishing the relationship between KC and stock returns, particularly in economic downturns, which serve as pivotal settings for examining corporate sustainability. Moreover, by identifying the moderating effects, our research offers insights into the conditions that amplify KC’s effectiveness during crisis periods. This comprehensive analysis not only elucidates the direct influence of KC but also strategically outlines conditions that maximize the utilization of KC. Our findings offer resource allocation strategies for corporate managers and help investment decision-making for investors seeking sustainable firms.

To understand how the results of our study manifest in the real world, utilizing difference-in-differences (DID) as a quasi-experimental research design, can help validate the robustness of our study. DID estimates analyze the difference in outcome changes before and after treatment between the treatment and control groups, thus providing insight into the impact of policies (Goodman-Bacon 2021). However, there are limitations in finding the policy effects, such as R&D tax incentives and subsidies affecting KC in specific regions, industries, or companies during the periods of the GFC and COVID-19. Therefore, DID for KC on crisis returns could be applied in countries other than the US, using policy implementation cases as future research opportunities.

However, since this empirical study was conducted on the U.S. manufacturing industry, there may be limitations in applying it to other countries and industries. Future research is needed to determine its applicability beyond the U.S. manufacturing sector. Furthermore, researching how KC correlates with the effects of general human capital, broadly argued as a key component of modern economic growth (Azarnert 2023), can enhance our understanding of intangible capital. Additionally, further study is needed to ascertain whether the effects of KC analyzed during GFC and COVID-19 are universally applicable by examining other economic crisis periods. Furthermore, there are opportunities for additional research on how the impact of KC on crisis returns may vary depending on the characteristics of economic crises.

Author Contributions

Conceptualization, C.C.L.; methodology, C.C.L. and H.K; software, C.C.L.; validation, C.C.L. and H.K.; formal analysis, C.C.L.; investigation, C.C.L.; resources, H.K; data curation, C.C.L.; writing—original draft preparation, C.C.L. and H.K.; writing—review and editing, E.A. and H.K.; visualization, C.C.L.; supervision, E.A. and H.K. All authors have read and agreed to the published version of the manuscript.

Funding

This work was supported by the Ministry of Education of the Republic of Korea and the National Research Foundation of Korea (NRF-2023S1A5A8081512).

Data Availability Statement

Data will be made available upon request.

Conflicts of Interest

The authors certify that there are no potential conflicts of interest relating to the subject matter discussed in this manuscript.

Appendix A. Variable Definitions and Sources

Variables	Description	Source
Dependent Variable
Daily Return	r – rf – 1 where r equals the daily return based on the price or bid/ask Average(PRC) in the CRSP database, and the rf is the 1-month daily Treasury bill rate	CRSP
Raw Return	$\prod (1 + d a i l y r e u t r n_{i, t}) - 1$ .	CRSP
Expected Return	Market model estimated daily return over the 60-month ending in the day before the onset of the crisis and normal period raw return on the four factors (market, book-to-market, size, and momentum)	CRSP and Kenneth French’s website⁽¹⁾
Abnormal Return	Raw Return—Expected return	CRSP
Main Independent Variable
Knowledge Capital to Total Assets	$K C_{i, t} = K C_{i, t - 1} (1 - δ_{0}) + (R & D_{i, t} / C P I_{t})$ . Initial stock: KC_i,t0 = R&D_i,t1/(g + δ₀) R&D represents the research and development expenses (XRD) CPI represents the consumer price index g = 10.08 percent δ₀ = 15 percent	COMPUSTAT
Crisis	A binary variable is set as one for the crisis period and zero otherwise.
Moderating Variables
Cash Holdings	Cash and cash equivalents (CHECH)_i,t/Total assets (AT)_i,t	COMPUSTAT
Market Share	Sales (SALE)_i,t/Total sales of an industry_i,t	COMPUSTAT
Tobin’s Q	{Common shares outstanding (CSHO)_i,t × Annual closing stock price (PRCC_F)_i,t + Long-term debt (DLTT)_i_,_t + Debt in current liabilities (DLC)_i_,_t}/Total assets (AT)_i_,_t Firms exceeding the industry-specific median each year are classified as higher firms of each moderator, while the rest are classified as lower firms.	COMPUSTAT
Control Variables
Market Capitalization	Log{Common shares outstanding (CSHO)_i_,_t × Annual closing stock price (PRCC_F)_i_,_t}	COMPUSTAT
Long-term Debt	Long-term debt (DLTT)_i_,_t/Total assets (AT)_i_,_t	COMPUSTAT
Short-term Debt	Debt in current liabilities (DLC)_i_,_t/Total assets (AT)_i_,_t	COMPUSTAT
Profitability	ROA = Net income (NI)_i_,_t/Total assets (AT)_i_,_t	COMPUSTAT
Book-to-Market	Common equity total (CEQ)_i_,_t/Market Cap.	COMPUSTAT
Negative B/M	The dummy variable is set as one if the book-to-market is negative and zero otherwise.	COMPUSTAT
Idiosyncratic Risk	The variance of the residuals in the market model’s expected return for sixty months leading up to the date preceding the onset of the crisis and its corresponding date for normal periods.	Kenneth French’s website
Fama–French Four-factor Loadings	Factor loadings of the Fama–French three factors plus the momentum factor—R_m-R_f, SMB, HML, and MOM—over the sixty months leading up to the date preceding the onset of the crisis and its corresponding date for normal periods.	Kenneth French’s website
Note: COMPUSTAT item codes are presented in parentheses. (1) https://mba.tuck.dartmouth.edu/pages/faculty/ken.french/data_library.html (accessed on 1 March 2024).

Appendix B. Descriptive Statistics for Non-Crisis and Crisis Periods

Panel A: Statistical Overview—Non-crisis periods in GFC
Variables	Observation	Mean	Std.Dev.	Min	25%	Median	75%	Max
KC	4849	0.233	0.552	0.002	0.040	0.105	0.245	5.311
Stock Return	4238	0.111	0.308	−0.821	−0.077	0.088	0.231	1.303
Abnormal Return	3516	−0.009	0.265	−0.702	−0.161	−0.031	−0.429	0.106
Market Cap.	4971	7.097	1.807	1.337	5.984	6.987	8.243	11.617
Long-term Debt	5498	0.154	0.183	0.000	0.000	0.103	0.242	1.159
Short-term Debt	5521	0.039	0.077	0.000	0.000	0.006	0.041	0.486
Cash Holdings	5507	0.062	0.082	0.000	0.012	0.033	0.077	0.481
Profitability	5512	0.112	0.115	−0.396	0.074	0.119	0.168	0.451
Book-to-Market	4970	0.586	0.586	−2.046	0.298	0.477	0.742	3.901
Negative B/M	5626	0.026	0.160	0.000	0.000	0.000	0.000	1.000
Momentum	4053	0.088	0.445	−0.746	−0.181	0.027	0.257	2.349
Idiosyncratic Risk	3543	0.031	0.019	0.011	0.021	0.026	0.034	0.154
Panel B: Statistical Overview—Crisis period in GFC
Variables	Observation	Mean	Std.Dev.	Min	25%	Median	75%	Max
KC	615	0.230	0.524	0.002	0.037	0.104	0.263	5.311
Stock Return	533	−0.556	0.202	−0.951	−0.700	−0.567	−0.429	0.106
Abnormal Return	429	−0.109	0.393	−0.898	−0.362	−0.136	0.094	1.353
Market Cap.	610	7.001	1.759	1.337	5.803	6.836	8.067	11.617
Long-term Debt	701	0.162	0.208	0.000	0.000	0.099	0.242	1.159
Short-term Debt	702	0.041	0.080	0.000	0.000	0.007	0.042	0.486
Cash Holdings	700	0.069	0.081	0.000	0.017	0.043	0.091	0.481
Profitability	701	0.085	0.115	−0.396	0.040	0.093	0.140	0.451
Book-to-Market	609	0.574	0.554	−2.046	0.346	0.521	0.777	3.901
Negative B/M	712	0.035	0.184	0.000	0.000	0.000	0.000	1.000
Momentum	515	−0.168	0.345	−0.746	−0.387	−0.180	−0.020	2.349
Idiosyncratic Risk	434	0.026	0.017	0.011	0.018	0.023	0.028	0.154
Panel C: Statistical Overview—Non-crisis periods in COVID-19
Variables	Observation	Mean	Std.Dev.	Min	25%	Median	75%	Max
KC	2438	0.089	0.199	−0.577	0.001	0.022	0.097	1.853
Stock Return	1699	−0.018	0.100	−0.530	−0.077	−0.019	0.035	0.301
Abnormal Return	1700	−0.008	0.099	−0.491	−0.061	−0.013	0.038	0.414
Market Cap.	2528	4.146	4.069	−2.382	0.000	5.207	7.816	12.332
Long-term Debt	2672	0.125	0.194	−0.270	0.000	0.040	0.220	1.084
Short-term Debt	2691	0.023	0.060	−0.113	0.000	0.005	0.021	0.417
Cash Holdings	2687	0.060	0.144	−0.274	−0.001	0.015	0.058	0.780
Profitability	2682	0.024	0.151	−0.647	−0.011	0.021	0.105	0.406
Book-to-Market	2510	0.530	0.940	−2.366	0.004	0.236	0.743	7.622
Negative B/M	2693	0.024	0.157	−0.750	0.000	0.000	0.000	1.000
Momentum	1702	0.011	0.322	−1.082	−0.176	−0.005	0.190	1.516
Idiosyncratic Risk	1702	0.007	0.011	−0.016	−0.001	0.001	0.016	0.045
Panel D: Statistical Overview—Crisis period in COVID-19
Variables	Observation	Mean	Std.Dev.	Min	25%	Median	75%	Max
KC	629	0.095	0.256	−0.259	−0.001	0.003	0.077	1.853
Stock Return	425	−0.403	0.137	−0.712	−0.488	−0.394	−0.311	0.045
Abnormal Return	426	−0.001	0.216	−0.491	−0.144	−0.005	0.150	0.414
Market Cap.	600	2.966	3.887	−0.796	0.043	0.260	7.065	11.555
Long-term Debt	684	0.098	0.174	−0.203	0.001	0.025	0.138	1.084
Short-term Debt	688	0.024	0.067	−0.082	0.000	0.004	0.019	0.417
Cash Holdings	687	0.070	0.151	−0.196	−0.000	0.021	0.074	0.780
Profitability	686	−0.004	0.139	−0.647	−0.028	−0.003	0.057	0.406
Book-to-Market	596	0.276	0.823	−2.481	−0.054	0.024	0.342	7.622
Negative B/M	696	0.018	0.140	−0.750	0.000	0.000	0.000	1.000
Momentum	426	−0.012	0.279	−0.975	−0.179	−0.015	0.155	1.107
Idiosyncratic Risk	426	0.002	0.007	−0.009	−0.001	−0.000	0.000	0.034

Appendix C. State-Level Cluster–Robust Standard Error Test

Panel A: GFC Period
	Raw Return			Abnormal Return
Variables	(1)	(2)	(3)	(4)	(5)	(6)
KC	0.012	0.010	0.010	0.009	0.004	0.009
	(0.013)	(0.014)	(0.050)	(0.016)	(0.014)	(0.058)
Crisis	0.089 ***	−0.095 **	−0.099 **	−0.046	−0.075 **	−0.079 ***
	(0.030)	(0.037)	(0.038)	(0.034)	(0.029)	(0.029)
KC×Crisis	0.094 **	0.120 **	0.107 **	0.075 **	0.098 **	0.096 **
	(0.043)	(0.051)	(0.053)	(0.037)	(0.046)	(0.042)
Market Cap.		−0.004	−0.066 ***		−0.003	−0.059 **
		(0.003)	(0.021)		(0.003)	(0.023)
Long-term Debt		0.052	−0.077 *		0.002	−0.110 **
		(0.037)	(0.045)		(0.040)	(0.047)
Short-term Debt		0.110	0.040		0.085	−0.024
		(0.129)	(0.150)		(0.125)	(0.167)
Cash Holdings		−0.114 *	−0.044		−0.092	−0.012
		(0.067)	(0.077)		(0.059)	(0.066)
Profitability		−0.184 **	−0.275		−0.222 ***	−0.264
		(0.081)	(0.178)		(0.076)	(0.188)
Book-to-Market		−0.040 ***	−0.110 ***		−0.062 ***	−0.138 ***
		(0.008)	(0.019)		(0.011)	(0.018)
Negative B/M		0.138	0.292 *		0.212 **	0.341 ***
		(0.119)	(0.151)		(0.087)	(0.114)
Momentum		−0.052 ***	−0.049 ***		−0.052 ***	−0.052 ***
		(0.010)	(0.013)		(0.009)	(0.012)
Idiosyncratic Risk		−2.594 ***	−4.107 ***		−1.670 ***	−2.453 ***
		(0.395)	(0.779)		(0.284)	(0.705)
Four-Factor Loadings	No	Yes	Yes	No	Yes	Yes
Industry Dummies	No	Yes	No	No	Yes	No
Firm Fixed Effects	No	No	Yes	No	No	Yes
N	3708	3062	3062	3708	3062	3062
Adj. R-Square	0.02	0.41	0.42	0.01	0.05	0.05
Panel B: COVID-19 Period
	Raw Return			Abnormal Return
Variables	(1)	(2)	(3)	(4)	(5)	(6)
KC	0.003	−0.053 **	−0.076	0.015	−0.066 **	−0.135
	(0.012)	(0.020)	(0.184)	(0.015)	(0.025)	(0.177)
Crisis	−0.401 ***	−0.195 ***	−0.185 ***	0.002	−0.246 ***	−0.227 ***
	(0.011)	(0.031)	(0.039)	(0.013)	(0.035)	(0.045)
KC×Crisis	0.133 ***	0.145 **	0.134 *	0.166 ***	0.227 **	0.209 *
	(0.046)	(0.061)	(0.067)	(0.063)	(0.099)	(0.110)
Market Cap.		−0.000	−0.063 ***		0.002	−0.076 ***
		(0.003)	(0.018)		(0.003)	(0.023)
Long-term Debt		−0.064 ***	0.028		−0.091 ***	0.035
		(0.019)	(0.070)		(0.024)	(0.076)
Short-term Debt		0.132	−0.071		0.125	−0.153
		(0.094)	(0.170)		(0.107)	(0.181)
Cash Holdings		0.000	−0.035		−0.004	−0.054
		(0.047)	(0.085)		(0.053)	(0.092)
Profitability		0.091 **	0.059		0.124 ***	0.008
		(0.035)	(0.130)		(0.042)	(0.153)
Book-to-Market		−0.016 ***	−0.022		−0.018 ***	−0.026
		(0.005)	(0.022)		(0.006)	(0.035)
Negative B/M		0.012	0.011		−0.010	−0.009
		(0.015)	(0.075)		(0.023)	(0.095)
Momentum		−0.014	0.003		−0.009	0.010
		(0.013)	(0.015)		(0.015)	(0.019)
Idiosyncratic Risk		−0.269	2.009		0.206	3.836
		(0.732)	(2.610)		(0.893)	(3.325)
Four-Factor Loadings	No	Yes	Yes	No	Yes	Yes
Industry Dummies	No	Yes	No	No	Yes	No
Firm Fixed Effects	No	No	Yes	No	No	Yes
N	1378	1072	1072	1378	1072	1072
Adj. R-Square	0.68	0.74	0.76	0.01	0.12	0.17
* p < 0.10, p < 0.05, and * p < 0.01.

References

Abadie, Alberto, Susan Athey, Guido W. Imbens, and Jeffrey M. Wooldridge. 2023. When should you adjust standard errors for clustering? The Quarterly Journal of Economics 138: 1–35. [Google Scholar] [CrossRef]
Afiezan, Adam, Giovanni Wijaya, and Cindy Claudia. 2020. The effect of free cash flow, company size, profitability and liquidity on debt policy for manufacturing companies listed on IDX in 2016–2019 periods. Budapest International Research and Critics Institute-Journal (BIRCI-Journal) 3: 4005–18. [Google Scholar] [CrossRef]
Ahmad, Manzoor, and Jianghuai Zheng. 2023. The cyclical and nonlinear impact of R&D and innovation activities on economic growth in OECD economies: A new perspective. Journal of the Knowledge Economy 14: 544–93. [Google Scholar]
Altelbany, Shady. 2021. Evaluation of ridge, elastic net and lasso regression methods in precedence of multicollinearity problem: A simulation study. Journal of Applied Economics and Business Studies 5: 131–42. [Google Scholar] [CrossRef]
Antonelli, Cristiano. 2019. Knowledge as an economic good: Exhaustibility versus appropriability? The Journal of Technology Transfer 44: 647–58. [Google Scholar] [CrossRef]
Antonelli, Cristiano, Gianluca Orsatti, and Guido Pialli. 2023. The knowledge-intensive direction of technological change. Eurasian Business Review 13: 1–27. [Google Scholar] [CrossRef]
Arellano, Manuel. 1987. Computing robust standard errors for within-groups estimators. Oxford Bulletin of Economics & Statistics 49: 431. [Google Scholar]
Azarnert, Leonid V. 2023. Population sorting and human capital accumulation. Oxford Economic Papers 75: 780–801. [Google Scholar] [CrossRef]
Bekaert, Geert, Michael Ehrmann, Marcel Fratzscher, and Arnaud Mehl. 2014. The global crisis and equity market contagion. The Journal of Finance 69: 2597–649. [Google Scholar] [CrossRef]
Bhattacharya, Abhi, Neil A. Morgan, and Lopo L. Rego. 2022. Examining why and when market share drives firm profit. Journal of Marketing 86: 73–94. [Google Scholar] [CrossRef]
Boehm, Christoph E., Aaron Flaaen, and Nitya Pandalai-Nayar. 2020. Multinationals, offshoring, and the decline of US manufacturing. Journal of International Economics 127: 103391. [Google Scholar] [CrossRef]
Bose, Sudipta, Syed Shams, Muhammad Jahangir Ali, and Dessalegn Mihret. 2022. COVID-19 impact, sustainability performance and firm value: International evidence. Accounting & Finance 62: 597–643. [Google Scholar]
Calomiris, Charles W., Inessa Love, and María Soledad Martínez Pería. 2012. Stock returns’ sensitivities to crisis shocks: Evidence from developed and emerging markets. Journal of International Money and Finance 31: 743–65. [Google Scholar] [CrossRef]
Carlin, Bruce Ian, Bhagwan Chowdhry, and Mark J. Garmaise. 2012. Investment in organization capital. Journal of Financial Intermediation 21: 268–86. [Google Scholar] [CrossRef]
Chan, Louis K. C., Josef Lakonishok, and Theodore Sougiannis. 2001. The stock market valuation of research and development expenditures. The Journal of Finance 56: 2431–56. [Google Scholar] [CrossRef]
Chang, Andrew C. 2018. Tax policy endogeneity: Evidence from R&D tax credits. Economics of Innovation and New Technology 27: 809–33. [Google Scholar]
Chang, Chong-Chuo, and Han Yang. 2022. The role of cash holdings during financial crises. Pacific-Basin Finance Journal 72: 101733. [Google Scholar] [CrossRef]
Chang, Jieun, and Youngho Kang. 2019. Instrumental variable estimates of the effect of management practices on firm performance in Korean firms. Journal of Labor Research 40: 106–25. [Google Scholar] [CrossRef]
Chen, Guoli, Sterling Huang, Philipp Meyer-Doyle, and Denisa Mindruta. 2021. Generalist versus specialist CEOs and acquisitions: Two-sided matching and the impact of CEO characteristics on firm outcomes. Strategic Management Journal 42: 1184–214. [Google Scholar] [CrossRef]
Chiu, Junmao, Chin-Ho Chen, Chung-Chieh Cheng, and Shih-Chang Hung. 2021. Knowledge capital, CEO power, and firm value: Evidence from the IT industry. The North American Journal of Economics and Finance 55: 101012. [Google Scholar] [CrossRef]
Coluccia, Daniela, Marina Dabić, Manlio Del Giudice, Stefano Fontana, and Silvia Solimene. 2020. R&D innovation indicator and its effects on the market. An empirical assessment from a financial perspective. Journal of Business Research 119: 259–71. [Google Scholar]
Corrado, Carol, Jonathan Haskel, Cecilia Jona-Lasinio, and Massimiliano Iommi. 2022. Intangible capital and modern economies. Journal of Economic Perspectives 36: 3–28. [Google Scholar] [CrossRef]
Cox, Josue, Daniel L. Greenwald, and Sydney C. Ludvigson. 2020. What Explains the COVID-19 Stock Market? Cambridge, MA: National Bureau of Economic Research. Available online: https://www.nber.org/papers/w27784 (accessed on 24 April 2024).
Cragg, John G., and Stephen G. Donald. 1993. Testing identifiability and specification in instrumental variable models. Econometric Theory 9: 222–40. [Google Scholar] [CrossRef]
Daniel, Kent, and Sheridan Titman. 1997. Evidence on the characteristics of cross sectional variation in stock returns. The Journal of Finance 52: 1–33. [Google Scholar] [CrossRef]
Dierickx, Ingemar, and Karel Cool. 1989. Asset stock accumulation and sustainability of competitive advantage. Management Science 35: 1504–11. [Google Scholar] [CrossRef]
Ding, Wenzhi, Ross Levine, Chen Lin, and Wensi Xie. 2021. Corporate immunity to the COVID-19 pandemic. Journal of Financial Economics 141: 802–30. [Google Scholar] [CrossRef]
Doraszelski, Ulrich, and Jordi Jaumandreu. 2013. R&D and productivity: Estimating endogenous productivity. Review of Economic Studies 80: 1338–83. [Google Scholar]
Eisfeldt, Andrea L., and Dimitris Papanikolaou. 2013. Organization capital and the cross-section of expected returns. The Journal of Finance 68: 1365–406. [Google Scholar] [CrossRef]
Eisfeldt, Andrea L., Edward Kim, and Dimitris Papanikolaou. 2020. Intangible Value. Cambridge, MA: National Bureau of Economic Research. Available online: https://www.nber.org/papers/w28056 (accessed on 24 April 2024).
Ewens, Michael, Ryan H. Peters, and Sean Wang. 2024. Measuring intangible capital with market prices. Management Science. [Google Scholar] [CrossRef]
Fahlenbrach, Rüdiger, Kevin Rageth, and René M. Stulz. 2021. How valuable is financial flexibility when revenue stops? Evidence from the COVID-19 crisis. The Review of Financial Studies 34: 5474–521. [Google Scholar] [CrossRef]
Fama, Eugene F., and Kenneth R. French. 1992. The cross-section of expected stock returns. The Journal of Finance 47: 427–65. [Google Scholar]
Faucheux, Sylvie. 2010. Intellectual and knowledge capital for sustainable development at local, national, regional, and global levels. Principles of Sustainable Development 2: 89–115. [Google Scholar]
Frankel, Jeffrey, and George Saravelos. 2012. Can leading indicators assess country vulnerability? Evidence from the 2008–09 global financial crisis. Journal of International Economics 87: 216–31. [Google Scholar] [CrossRef]
Goodman-Bacon, Andrew. 2021. Difference-in-differences with variation in treatment timing. Journal of econometrics 225: 254–77. [Google Scholar] [CrossRef]
Gu, Lifeng. 2016. Product market competition, R&D investment, and stock returns. Journal of Financial Economics 119: 441–55. [Google Scholar]
Habib, Shafaat Muhammad, Haroon Hussain, Mamdouh Abdulaziz Saleh Al-Faryan, and Rana Yassir Hussain. 2022. Impact of firm characteristics and ownership structure on firm efficiency: Evidence from non-financial firms of Pakistan. Cogent Economics & Finance 10: 2106628. [Google Scholar]
Hansen, Lars Peter. 1982. Large sample properties of generalized method of moments estimators. Econometrica: Journal of the Econometric Society, 1029–54. [Google Scholar]
Harford, Jarrad, Sandy Klasa, and William F. Maxwell. 2014. Refinancing risk and cash holdings. The Journal of Finance 69: 975–1012. [Google Scholar] [CrossRef]
Hartzmark, Samuel M., and Abigail B. Sussman. 2019. Do investors value sustainability? A. natural experiment examining ranking and fund flows. The Journal of Finance 74: 2789–837. [Google Scholar] [CrossRef]
Hasan, Mostafa Monzur. 2018. Organization capital and firm life cycle. Journal of corporate finance 48: 556–78. [Google Scholar] [CrossRef]
Hasan, Mostafa Monzur, and Mohammad Riaz Uddin. 2022. Do intangibles matter for corporate policies? Evidence from organization capital and corporate payout choices. Journal of Banking & Finance 135: 106395. [Google Scholar]
Hasan, Mostafa Monzur, Gerald J. Lobo, and Buhui Qiu. 2021. Organizational capital, corporate tax avoidance, and firm value. Journal of Corporate Finance 70: 102050. [Google Scholar] [CrossRef]
Hausman, Jerry A. 1978. Specification tests in econometrics. Econometrica: Journal of the Econometric Society, 1251–71. [Google Scholar] [CrossRef]
He, Zhaozhao, and M. Babajide Wintoki. 2016. The cost of innovation: R&D and high cash holdings in US firms. Journal of Corporate Finance 41: 280–303. [Google Scholar]
Hegde, Shantaram P., and Dev R. Mishra. 2023. Patented knowledge capital and implied equity risk premium. Journal of Banking & Finance 148: 106738. [Google Scholar]
Hombert, Johan, and Adrien Matray. 2018. Can innovation help US manufacturing firms escape import competition from China? The Journal of Finance 73: 2003–39. [Google Scholar] [CrossRef]
Huang, Qianqian, and Tao Yuan. 2021. Does political corruption impede firm innovation? Evidence from the United States. Journal of Financial and Quantitative Analysis 56: 213–48. [Google Scholar] [CrossRef]
Hula, David G. 1989. Intangible capital, market share and corporate strategy. Applied Economics 21: 1535–47. [Google Scholar] [CrossRef]
Jan, Ching-Lih, and Jane A. Ou. 2020. The Information Content of Negative Book Value. Journal of Accounting & Finance (2158–3625) 20. [Google Scholar]
Kim, JooMan, Insun Yang, Taeyong Yang, and Peter Koveos. 2021. The impact of R&D intensity, financial constraints, and dividend payout policy on firm value. Finance Research Letters 40: 101802. [Google Scholar]
Krugman, Paul R. 1997. Development, Geography, and Economic Theory. Cambridge, MA: MIT Press, vol. 6. [Google Scholar]
Laperche, Blandine. 2021. Large firms’ knowledge capital and innovation networks. Journal of the Knowledge Economy 12: 183–200. [Google Scholar] [CrossRef]
Lewbel, Arthur. 2012. Using heteroscedasticity to identify and estimate mismeasured and endogenous regressor models. Journal of Business & Economic Statistics 30: 67–80. [Google Scholar]
Li, Guo, Na Li, and Suresh P. Sethi. 2021. Does CSR reduce idiosyncratic risk? Roles of operational efficiency and AI innovation. Production and Operations Management 30: 2027–45. [Google Scholar] [CrossRef]
Lins, Karl V., Henri Servaes, and Ane Tamayo. 2017. Social capital, trust, and firm performance: The value of corporate social responsibility during the financial crisis. The Journal of Finance 72: 1785–824. [Google Scholar] [CrossRef]
Lome, Ola, Alf Gunnar Heggeseth, and Øystein Moen. 2016. The effect of R&D on performance: Do R&D-intensive firms handle a financial crisis better? The Journal of High Technology Management Research 27: 65–77. [Google Scholar]
MacKinnon, James G., Morten Ørregaard Nielsen, and Matthew D. Webb. 2023. Testing for the appropriate level of clustering in linear regression models. Journal of Econometrics 235: 2027–56. [Google Scholar] [CrossRef]
Mavis, Christos P., Nathan P. McNamee, Dimitris Petmezas, and Nickolaos G. Travlos. 2020. Selling to buy: Asset sales and acquisitions. Journal of Corporate Finance 62: 101587. [Google Scholar] [CrossRef]
McNeish, Daniel M. 2015. Using lasso for predictor selection and to assuage overfitting: A method long overlooked in behavioral sciences. Multivariate Behavioral Research 50: 471–84. [Google Scholar] [CrossRef]
Nugroho, B. Yuliarto. 2020. The effect of book to market ratio, profitability, and investment on stock return. International Journal of Economics and Management Studies 7: 102–7. [Google Scholar]
Parcharidis, Efstathios G., and Nikos C. Varsakelis. 2010. R&D and Tobin’s q in an emerging financial market: The case of the Athens Stock Exchange. Managerial and Decision Economics 31: 353–61. [Google Scholar]
Pereira, Vijay, and Umesh Bamel. 2021. Extending the resource and knowledge based view: A critical analysis into its theoretical evolution and future research directions. Journal of Business Research 132: 557–70. [Google Scholar] [CrossRef]
Peters, Bettina, Mark J. Roberts, Van Anh Vuong, and Helmut Fryges. 2017. Estimating dynamic R&D choice: An analysis of costs and long-run benefits. The RAND Journal of Economics 48: 409–37. [Google Scholar]
Peters, Ryan H., and Lucian A. Taylor. 2017. Intangible capital and the investment-q relation. Journal of Financial Economics 123: 251–72. [Google Scholar] [CrossRef]
Roper, Stephen, and Joanne Turner. 2020. R&D and innovation after COVID-19: What can we expect? A review of prior research and data trends after the great financial crisis. International Small Business Journal 38: 504–14. [Google Scholar]
Sanford, Anthony, and Mu-Jeung Yang. 2022. Corporate investment and growth opportunities: The role of R&D-capital complementarity. Journal of Corporate Finance 72: 102130. [Google Scholar]
Sha, Yezhou, Ping Zhang, Yiru Wang, and Yifan Xu. 2022. Capital market opening and green innovation——Evidence from Shanghai-Hong Kong stock connect and the Shenzhen-Hong Kong stock connect. Energy Economics 111: 106048. [Google Scholar] [CrossRef]
Sheng, Dachen, and Heather Montgomery. 2022. Should Firms in Emerging Markets Invest in R&D? Evidence from China’s Manufacturing Sector. Journal of Risk and Financial Management 15: 517. [Google Scholar]
Stock, James, and Motohiro Yogo. 2005. Asymptotic distributions of instrumental variables statistics with many instruments. Identification and Inference for Econometric Models: Essays in Honor of Thomas Rothenberg 6: 109–20. [Google Scholar]
Trinh, Vu Quang, Ngan Duong Cao, Teng Li, and Marwa Elnahass. 2023. Social capital, trust, and bank tail risk: The value of ESG rating and the effects of crisis shocks. Journal of International Financial Markets, Institutions and Money 83: 101740. [Google Scholar] [CrossRef]
Ugur, Mehmet, and Marco Vivarelli. 2021. Innovation, firm survival and productivity: The state of the art. Economics of Innovation and New Technology 30: 433–67. [Google Scholar] [CrossRef]
Ullah, Subhan, Pervaiz Akhtar, and Ghasem Zaefarian. 2018. Dealing with endogeneity bias: The generalized method of moments (GMM) for panel data. Industrial Marketing Management 71: 69–78. [Google Scholar] [CrossRef]
White, Halbert. 1980. A heteroskedasticity-consistent covariance matrix estimator and a direct test for heteroskedasticity. Econometrica: Journal of the Econometric Society, 817–38. [Google Scholar] [CrossRef]
Willekens, Marleen, Simon Dekeyser, Liesbeth Bruynseels, and Wieteke Numan. 2023. Auditor market power and audit quality revisited: Effects of market concentration, market share distance, and leadership. Journal of Accounting, Auditing & Finance 38: 161–81. [Google Scholar]
Wright, Stephen. 2004. Measures of stock market value and returns for the US nonfinancial corporate sector, 1900–2002. Review of Income and Wealth 50: 561–84. [Google Scholar] [CrossRef]
Yoo, Sunbin, Alexander Ryota Keeley, and Shunsuke Managi. 2021. Does sustainability activities performance matter during financial crises? Investigating the case of COVID-19. Energy Policy 155: 112330. [Google Scholar] [CrossRef]
Zaefarian, Ghasem, Vita Kadile, Stephan C. Henneberg, and Alexander Leischnig. 2017. Endogeneity bias in marketing research: Problem, causes and remedies. Industrial Marketing Management 65: 39–46. [Google Scholar] [CrossRef]
Zhang, Ping, Yiru Wang, Ruyan Wang, and Tewei Wang. 2024. Digital finance and corporate innovation: Evidence from China. Applied Economics 56: 615–38. [Google Scholar] [CrossRef]
Zhang, Yaojie, Feng Ma, and Yudong Wang. 2019. Forecasting crude oil prices with a large set of predictors: Can LASSO select powerful predictors? Journal of Empirical Finance 54: 97–117. [Google Scholar] [CrossRef]

Table 1. Descriptive statistics and correlations.

Panel A: Statistical Overview—GFC
Variables		Observation		Mean		Std.Dev.		Min		25%		Median			75%		Max
KC		5464		0.232		0.549		0.002		0.039		0.105			0.245		5.311
Raw Return		4771		0.037		0.363		−0.821		−0.158		0.050			0.231		1.303
Abnormal Return		3945		−0.019		0.280		−0.702		−0.177		−0.038			0.111		1.082
Crisis Raw Return		533		−0.556		0.202		−0.951		−0.700		−0.567			−0.429		0.106
Crisis Abn. Return		429		−0.109		0.393		−0.898		−0.362		−0.136			0.094		1.353
Market Cap.		5581		7.087		1.802		1.337		5.964		6.971			8.232		11.617
Long-term Debt		6199		0.155		0.186		0.000		0.000		0.103			0.242		1.159
Short-term Debt		6223		0.039		0.077		0.000		0.000		0.006			0.041		0.486
Cash Holdings		6207		0.062		0.082		0.000		0.012		0.034			0.079		0.481
Profitability		6213		0.109		0.115		−0.396		0.070		0.116			0.166		0.451
Book-to-Market		5579		0.585		0.583		−2.046		0.302		0.481			0.747		3.901
Negative B/M		6338		0.027		0.162		0.000		0.000		0.000			0.000		1.000
Momentum		4568		0.059		0.442		−0.746		−0.207		−0.003			0.229		2.349
Idiosyncratic Risk		3977		0.030		0.019		0.011		0.020		0.026			0.034		0.154
Panel B: Statistical Overview—COVID-19
Variables		Observation		Mean		Std.Dev.		Min		25%		Median			75%		Max
KC		2688		0.145		0.214		0.002		0.033		0.075			0.165		1.380
Raw Return		2128		−0.095		0.188		−0.712		−0.152		−0.045			0.021		0.301
Abnormal Return		2128		−0.007		0.131		−0.491		−0.069		−0.012			0.049		0.414
Crisis Raw Return		423		−0.402		0.130		−0.650		−0.489		−0.394			−0.312		−0.027
Crisis Abn. Return		423		−0.058		0.249		−0.620		−0.224		−0.054			0.106		0.599
Market Cap.		3128		8.019		1.789		2.506		6.802		7.962			9.240		12.332
Long-term Debt		3356		0.225		0.191		0.000		0.061		0.204			0.328		1.084
Short-term Debt		3379		0.036		0.064		0.000		0.004		0.012			0.037		0.417
Cash Holdings		3374		0.085		0.141		0.000		0.012		0.034			0.085		0.780
Profitability		3368		0.072		0.163		−0.647		0.052		0.103			0.148		0.406
Book-to-Market		3106		0.841		0.945		−0.114		0.314		0.567			1.011		7.622
Negative B/M		3389		0.037		0.189		0.000		0.000		0.000			0.000		1.000
Momentum		2128		0.128		0.368		−0.608		−0.110		0.085			0.300		1.516
Idiosyncratic Risk		2128		0.020		0.008		0.008		0.014		0.018			0.024		0.045
Panel C: Correlataion Matrix—GFC
Variables	(1)		(2)	(3)	(4)		(5)		(6)		(7)	(8)	(9)			(10)		(11)
KC (1)
Raw Return (2)	0.030 **
Abnormal Return (3)	0.048 *		0.661 ***
Crisis Raw Return (4)	0.071 ***		1.000 ***	0.847 ***
Crisis Abn. Return (5)	−0.117 **		0.838 ***	1.000 ***	0.816 ***
Market Cap (6)	−0.292 ***		0.044 ***	0.031 *	0.311 ***		0.210 ***
Long-term Debt (7)	−0.098 ***		0.013	−0.004	−0.216 ***		−0.193 ***		0.078 ***
Short-term Debt (8)	−0.029 **		−0.032 **	−0.024	−0.166 ***		−0.120 **		−0.132 ***		0.001
Cash Holdings (9)	0.153 ***		−0.036 **	−0.014	0.112 ***		0.104 **		−0.163 ***		−0.198 ***	−0.068 ***
Profitability (10)	−0.281 ***		0.034 **	0.003	0.160 ***		0.082 *		0.369 ***		0.038 ***	−0.101 ***	−0.192 ***
Book-to-Market (11)	−0.122 ***		−0.187 ***	−0.065 ***	−0.321 ***		−0.180 ***		−0.215 ***		−0.151 ***	0.081 ***	−0.053 ***			−0.169 ***
Negative B/M (12)	0.125 ***		−0.016	−0.032 **	−0.074		−0.097 **		−0.192 ***		0.262 ***	0.158 ***	−0.020 *			−0.096 ***		−0.407 ***
Panel D: Correlataion Matrix—COVID-19
Variables	(1)		(2)	(3)	(4)		(5)		(6)		(7)	(8)		(9)		(10)		(11)
KC (1)
Stock Return (2)	0.052 **
Crisis Return (3)	0.048 *		0.449 ***
Market Cap (4)	0.185 ***		1.000 ***	0.728 ***
Long-term Debt (5)	0.143 ***		0.733 ***	1.000 ***	0.728 ***
Short-term Debt (6)	−0.180 ***		0.031	0.053 **	0.204 ***		0.113 **
Cash Holdings (7)	−0.162 ***		−0.085 ***	−0.096 ***	−0.396 ***		−0.263 ***		0.089 ***
Profitability (8)	0.084 ***		−0.001	0.014	−0.061		−0.047		0.029		−0.050 ***
Book-to-Market (9)	0.338 ***		−0.052 **	0.003	0.051		−0.016		−0.266 ***		−0.167 ***	0.028
Negative B/M (10)	−0.509 ***		0.127 ***	0.089 ***	0.311 ***		0.227 ***		0.429 ***		0.107 ***	−0.114 ***		−0.525 ***
Momentum (11)	−0.147 ***		−0.055 **	−0.074 ***	−0.299 ***		−0.112 **		−0.207 ***		−0.001	0.210 ***		−0.100 ***		−0.105 ***
Idiosync. Risk (12)	0.023		−0.031	−0.055 **	−0.131 ***		−0.143 ***		−0.039 **		0.383 ***	0.113 ***		−0.004		−0.008		−0.011

* p < 0.10, ** p < 0.05, and *** p < 0.01.

Table 2. KC and stock returns in GFC.

	Raw Return			Abnormal Return
Variables	(1)	(2)	(3)	(4)	(5)	(6)
KC	0.017	0.003	0.008	−0.002	−0.007	0.009
	(0.016)	(0.015)	(0.037)	(0.014)	(0.015)	(0.036)
Crisis	0.086 ***	−0.100 ***	−0.101 ***	−0.057 **	−0.084 ***	−0.086 ***
	(0.029)	(0.025)	(0.026)	(0.025)	(0.025)	(0.026)
KC×Crisis	0.085 **	0.128 ***	0.134 ***	0.092 **	0.112 ***	0.108 ***
	(0.041)	(0.037)	(0.051)	(0.038)	(0.037)	(0.037)
Market Cap.		−0.006 *	−0.065 ***		−0.003	−0.057 ***
		(0.003)	(0.018)		(0.003)	(0.018)
Long-term Debt		0.053	−0.091		−0.008	−0.125 **
		(0.033)	(0.063)		(0.033)	(0.063)
Short-term Debt		0.119	−0.004		0.118	−0.109
		(0.097)	(0.138)		(0.096)	(0.138)
Cash Holdings		−0.078	−0.034		−0.046	0.014
		(0.076)	(0.088)		(0.075)	(0.088)
Profitability		−0.129 **	−0.205 **		−0.152 **	−0.205 **
		(0.061)	(0.102)		(0.060)	(0.102)
Book-to-Market		−0.033 ***	−0.090 ***		−0.047 ***	−0.113 ***
		(0.007)	(0.012)		(0.007)	(0.012)
Negative B/M		0.119	0.293 *		0.206	0.342 *
		(0.160)	(0.176)		(0.158)	(0.175)
Momentum		−0.049 ***	−0.049 ***		−0.050 ***	−0.051 ***
		(0.010)	(0.011)		(0.010)	(0.011)
Idiosyncratic Risk		−2.889 ***	−4.098 ***		−1.788 ***	−2.559 ***
		(0.297)	(0.446)		(0.293)	(0.445)
Four-Factor Loadings	No	Yes	Yes	No	Yes	Yes
Industry Dummies	No	Yes	No	No	Yes	No
Firm Fixed Effects	No	No	Yes	No	No	Yes
N	4415	3484	3484	4415	3484	3484
Adj. R-Square	0.03	0.40	0.42	0.02	0.05	0.07

* p < 0.10, ** p < 0.05, and *** p < 0.01.

Table 3. KC and stock returns in COVID-19.

	Raw Return			Abnormal Return
Variables	(1)	(2)	(3)	(4)	(5)	(6)
KC	−0.002	−0.051 **	−0.052	0.010	−0.063 **	−0.107
	(0.019)	(0.025)	(0.140)	(0.023)	(0.031)	(0.177)
Crisis	−0.397 ***	−0.172 ***	−0.157 ***	0.004	−0.223 ***	−0.199 ***
	(0.008)	(0.022)	(0.024)	(0.010)	(0.028)	(0.030)
KC×Crisis	0.151 ***	0.165 ***	0.158 ***	0.186 ***	0.252 ***	0.241 ***
	(0.043)	(0.043)	(0.042)	(0.051)	(0.054)	(0.054)
Market Cap.		0.000	−0.057 ***		0.002	−0.068 ***
		(0.002)	(0.012)		(0.003)	(0.015)
Long-term Debt		−0.060 ***	0.034		−0.081 ***	0.052
		(0.021)	(0.060)		(0.026)	(0.076)
Short-term Debt		0.112 *	−0.095		0.100	−0.167
		(0.065)	(0.130)		(0.081)	(0.164)
Cash Holdings		0.029	−0.012		0.028	−0.034
		(0.046)	(0.059)		(0.057)	(0.075)
Profitability		0.094 **	0.155		0.123 **	0.123
		(0.041)	(0.094)		(0.051)	(0.119)
Book-to-Market		−0.010 **	−0.006		−0.013 **	−0.010
		(0.004)	(0.012)		(0.006)	(0.016)
Negative B/M		0.011	−0.001		−0.014	−0.022
		(0.021)	(0.054)		(0.027)	(0.068)
Momentum		−0.018 **	−0.004		−0.014	0.001
		(0.009)	(0.011)		(0.011)	(0.013)
Idiosyncratic Risk		−0.479	2.263 *		−0.054	3.971 **
		(0.518)	(1.334)		(0.648)	(1.686)
Four-Factor Loadings	No	Yes	Yes	No	Yes	Yes
Industry Dummies	No	Yes	No	No	Yes	No
Firm Fixed Effects	No	No	Yes	No	No	Yes
N	1625	1528	1528	1625	1528	1528
Adj. R-Square	0.67	0.73	0.74	0.01	0.11	0.16

* p < 0.10, ** p < 0.05, and *** p < 0.01.

Table 4. Moderating role of market share, Tobin’s Q, and cash holdings.

Panel A: Moderating Effect Test—GFC
Dependent Variable: Raw Return
Variables	High Market Share	Low Market Share	High Tobin’s Q	Low Tobin’s Q	High Cash Hold.	Low Cash Hold.
Variables	(1)	(2)	(3)	(4)	(5)	(6)
KC	0.010	0.041	0.074	0.064	−0.042	0.196 ***
	(0.065)	(0.055)	(0.061)	(0.065)	(0.055)	(0.064)
Crisis	−0.096 ***	−0.138 ***	−0.104 ***	−0.079 **	−0.104 **	−0.067 *
	(0.034)	(0.045)	(0.040)	(0.035)	(0.044)	(0.034)
KC×Crisis	0.149 ***	0.074	0.171 ***	0.081	0.202 ***	−0.017
	(0.057)	(0.054)	(0.059)	(0.052)	(0.065)	(0.050)
Market Cap.	−0.068 ***	−0.072 **	−0.043	−0.085 ***	−0.089 ***	−0.016
	(0.024)	(0.030)	(0.034)	(0.028)	(0.028)	(0.027)
Long-term Debt	−0.150 *	−0.003	−0.041	−0.009	−0.043	−0.115
	(0.082)	(0.111)	(0.110)	(0.087)	(0.101)	(0.089)
Short-term Debt	−0.019	−0.002	−0.262	0.074	0.126	−0.285
	(0.180)	(0.221)	(0.202)	(0.204)	(0.215)	(0.205)
Cash Holdings	0.053	−0.065	0.049	0.103	−0.105	0.139
	(0.134)	(0.125)	(0.149)	(0.113)	(0.133)	(0.612)
Profitability	−0.206	−0.009	−0.153	−0.287 **	−0.145	−0.199
	(0.151)	(0.156)	(0.178)	(0.133)	(0.149)	(0.172)
Book-to-Market	−0.077 ***	−0.122 ***	−0.070 ***	−0.270 ***	−0.128 ***	−0.045 ***
	(0.016)	(0.021)	(0.017)	(0.054)	(0.019)	(0.017)
Negative B/M	0.211	0.456	0.291	0.190	0.352	0.102
	(0.204)	(0.317)	(0.174)	(0.172)	(0.235)	(0.280)
Momentum	−0.042 ***	−0.044 ***	−0.061 ***	−0.054 ***	−0.036 **	−0.068 ***
	(0.015)	(0.017)	(0.017)	(0.015)	(0.016)	(0.017)
Idiosyncratic Risk	−3.616 ***	−4.891 ***	−3.205 ***	−5.292 ***	−5.309 ***	−2.068 ***
	(0.557)	(0.766)	(0.655)	(0.735)	(0.689)	(0.680)
Four-factor Loadings	Yes	Yes	Yes	Yes	Yes	Yes
Firm Fixed Effects	Yes	Yes	Yes	Yes	Yes	Yes
N	1999	1485	1777	1707	1671	1813
Adj. R-squared	0.46	0.40	0.41	0.45	0.40	0.48
Panel B: Moderating Effect Test—GFC
Dependent Variable: Abnormal Return
Variables	High Market Share	Low Market Share	High Tobin’s Q	Low Tobin’s Q	High Cash Hold.	Low Cash Hold.
Variables	(1)	(2)	(3)	(4)	(5)	(6)
KC	0.070	0.005	0.081	0.093	−0.021	0.182 ***
	(0.055)	(0.065)	(0.062)	(0.064)	(0.055)	(0.064)
Crisis	−0.115 **	−0.091 ***	−0.110 ***	−0.098 ***	−0.071	−0.052
	(0.045)	(0.034)	(0.040)	(0.035)	(0.045)	(0.034)
KC×Crisis	0.146 ***	0.087	0.167 ***	0.079	0.164 **	−0.015
	(0.054)	(0.062)	(0.059)	(0.051)	(0.066)	(0.050)
Market Cap.	−0.054 *	−0.060 **	−0.022	−0.061 **	−0.080 ***	−0.013
	(0.030)	(0.024)	(0.034)	(0.028)	(0.028)	(0.026)
Long-term Debt	−0.065	−0.157 *	−0.117	0.031	−0.063	−0.161 *
	(0.111)	(0.082)	(0.111)	(0.087)	(0.102)	(0.089)
Short-term Debt	−0.197	−0.040	−0.273	0.124	−0.052	−0.343 *
	(0.220)	(0.180)	(0.204)	(0.203)	(0.216)	(0.204)
Cash Holdings	−0.009	0.103	0.074	0.038	−0.089	0.141
	(0.125)	(0.134)	(0.152)	(0.114)	(0.133)	(0.610)
Profitability	0.061	−0.310 **	−0.162	−0.300 **	−0.105	−0.247
	(0.156)	(0.151)	(0.183)	(0.135)	(0.149)	(0.171)
Book-to-Market	−0.139 ***	−0.102 ***	−0.095 ***	−0.246 ***	−0.153 ***	−0.066 ***
	(0.020)	(0.016)	(0.017)	(0.055)	(0.019)	(0.017)
Negative B/M	0.447	0.284	0.287	0.193	0.346	0.220
	(0.316)	(0.204)	(0.171)	(0.167)	(0.236)	(0.279)
Momentum	−0.046 ***	−0.043 ***	−0.056 ***	−0.052 ***	−0.040 **	−0.067 ***
	(0.017)	(0.015)	(0.018)	(0.015)	(0.017)	(0.017)
Idiosyncratic Risk	−3.105 ***	−2.285 ***	−1.965 ***	−2.914 ***	−3.489 ***	−0.999
	(0.764)	(0.557)	(0.639)	(0.717)	(0.692)	(0.678)
Four-factor Loadings	Yes	Yes	Yes	Yes	Yes	Yes
Firm Fixed Effects	Yes	Yes	Yes	Yes	Yes	Yes
N	1999	1485	1777	1707	1671	1813
Adj. R-squared	0.08	0.07	0.08	0.12	0.10	0.08
Panel C: Moderating Effect Test—COVID-19
Dependent Variable: Raw Return
Variables	High Market Share	Low Market Share	High Tobin’s Q	Low Tobin’s Q	High Cash Hold.	Low Cash Hold.
Variables	(1)	(2)	(3)	(4)	(5)	(6)
KC	0.099	−0.064	0.296	0.009	−0.210	−0.114
	(0.205)	(0.136)	(0.368)	(0.114)	(0.198)	(0.154)
Crisis	−0.172 ***	−0.155 ***	−0.155 ***	−0.187 ***	−0.185 ***	−0.125 ***
	(0.026)	(0.036)	(0.035)	(0.029)	(0.029)	(0.041)
KC×Crisis	0.442 ***	0.025	0.259 **	0.047	0.389 ***	0.054
	(0.074)	(0.044)	(0.104)	(0.039)	(0.095)	(0.047)
Market Cap.	−0.030 **	−0.084 ***	−0.018	−0.059 ***	−0.052 ***	−0.067 ***
	(0.014)	(0.018)	(0.022)	(0.018)	(0.015)	(0.018)
Long-term Debt	−0.060	0.059	0.111	0.042	−0.057	0.149
	(0.077)	(0.093)	(0.106)	(0.072)	(0.088)	(0.099)
Short-term Debt	−0.090	0.117	0.100	−0.066	−0.224	0.243
	(0.149)	(0.186)	(0.208)	(0.145)	(0.199)	(0.185)
Cash Holdings	0.063	0.034	0.060	0.020	−0.647 *	0.047
	(0.072)	(0.083)	(0.118)	(0.061)	(0.367)	(0.084)
Profitability	0.052	0.218 *	0.413 ***	0.162	0.202	0.223
	(0.117)	(0.129)	(0.158)	(0.111)	(0.135)	(0.140)
Book-to-Market	0.020 *	−0.046	0.024 *	0.056	0.018	−0.052 *
	(0.010)	(0.029)	(0.013)	(0.077)	(0.011)	(0.030)
Negative B/M	−0.015	0.146	−0.002	−0.085	0.067	−0.113 *
	(0.044)	(0.115)	(0.080)	(0.061)	(0.083)	(0.068)
Momentum	−0.011	−0.001	−0.002	0.010	−0.008	−0.013
	(0.013)	(0.015)	(0.017)	(0.012)	(0.015)	(0.016)
Idiosyncratic Risk	1.068	5.010 **	−0.702	4.455 **	1.698	2.325
	(1.327)	(2.200)	(1.689)	(1.810)	(1.524)	(2.502)
Four-factor Loadings	Yes	Yes	Yes	Yes	Yes	Yes
Firm Fixed Effects	Yes	Yes	Yes	Yes	Yes	Yes
N	910	618	722	806	859	669
Adj. R-squared	0.79	0.69	0.77	0.73	0.81	0.65
Panel D: Moderating Effect Test—COVID-19
Dependent Variable: Abnormal Return
Variables	High Market Share	Low Market Share	High Tobin’s Q	Low Tobin’s Q	High Cash Hold.	Low Cash Hold.
Variables	(1)	(2)	(3)	(4)	(5)	(6)
KC	0.153	−0.498 *	−0.238	0.631	−0.326	−0.422
	(0.264)	(0.275)	(0.193)	(0.480)	(0.271)	(0.312)
Crisis	−0.235 ***	−0.203 ***	−0.247 ***	−0.220 ***	−0.254 ***	−0.123 **
	(0.036)	(0.051)	(0.040)	(0.048)	(0.041)	(0.058)
KC×Crisis	0.721 ***	0.067	0.130 **	0.054	0.478 ***	0.081
	(0.101)	(0.069)	(0.060)	(0.076)	(0.130)	(0.072)
Market Cap.	−0.040 **	−0.130 ***	−0.088 ***	−0.020	−0.102 ***	−0.095 ***
	(0.020)	(0.027)	(0.020)	(0.029)	(0.023)	(0.025)
Long-term Debt	0.010	0.030	−0.001	0.209	0.005	0.232 *
	(0.101)	(0.134)	(0.100)	(0.141)	(0.121)	(0.136)
Short-term Debt	−0.118	−0.051	−0.171	−0.056	−0.273	0.113
	(0.225)	(0.253)	(0.191)	(0.313)	(0.279)	(0.259)
Cash Holdings	0.049	−0.045	−0.035	0.073	−0.922 *	0.019
	(0.101)	(0.116)	(0.081)	(0.161)	(0.510)	(0.117)
Profitability	0.047	0.104	0.018	0.323	0.155	0.258
	(0.166)	(0.190)	(0.160)	(0.207)	(0.198)	(0.189)
Book-to-Market	0.009	−0.084 **	−0.016	0.031	−0.011	−0.085 **
	(0.017)	(0.037)	(0.073)	(0.021)	(0.020)	(0.037)
Negative B/M	−0.005	0.200	−0.054	−0.075	0.041	−0.150
	(0.065)	(0.146)	(0.080)	(0.122)	(0.106)	(0.106)
Momentum	0.001	0.007	0.015	−0.019	0.008	0.002
	(0.018)	(0.022)	(0.017)	(0.022)	(0.021)	(0.023)
Idiosyncratic Risk	2.488	6.019 **	5.539 **	5.859 **	4.924 **	4.444
	(2.033)	(3.048)	(2.584)	(2.506)	(2.289)	(3.518)
Four-factor Loadings	Yes	Yes	Yes	Yes	Yes	Yes
Firm Fixed Effects	Yes	Yes	Yes	Yes	Yes	Yes
N	910	618	722	806	859	669
Adj. R-squared	0.20	0.19	0.26	0.18	0.24	0.11

* p < 0.10, ** p < 0.05, and *** p < 0.01.

Table 5. KC and crises returns: Two-stage instrumental variable estimation.

Panel A: Two-Stage Least Square Test for GFC Period
Variables	1st Stage	2nd Stage		Lewbel
	(1)	(2)	(3)	(4)	(3)
	KC	Raw Returns	Abn. Return	Raw Returns	Abn. Return
KC		0.073 **	0.089 **	0.007	0.013
		(0.035)	(0.044	(0.050)	(0.047)
Crisis		−0.024 ***	−0.027 **	−0.101 ***	−0.087 ***
		(0.009)	(0.012)	(0.027)	(0.022)
KC×Crisis		0.119 ***	0.188 ***	0.109 ***	0.105 ***
		(0.044)	(0.066)	(0.038)	(0.039)
Ind. Mean KC (Inst.Var.)	0.041 **
	(0.016)
2 Years Lag KC (Inst. Var.)	0.318 ***
	(0.019)
State Mean KC (Inst. Var.)	0.037 ***
	(0.014)
Market Cap.	−0.278 ***	0.005	0.009 **	−0.063 ***	−0.056 **
	(0.009)	(0.003)	(0.004)	(0.024)	(0.023)
Long-term Debt	−0.378 ***	−0.079 ***	−0.113 ***	−0.090	−0.122 *
	(0.036)	(0.026)	(0.034)	(0.069)	(0.072)
Short-term Debt	0.238 ***	0.068	0.044	0.016	−0.078
	(0.077)	(0.099)	(0.130)	(0.142)	(0.150)
Cash Holdings	−0.014	−0.061	−0.078	−0.028	0.015
	(0.047)	(0.059)	(0.077)	(0.098)	(0.094)
Profitability	−0.061	0.187 ***	0.254 ***	−0.212	−0.213 *
	(0.056)	(0.054)	(0.070)	(0.130)	(0.119)
Book-to-Market	−0.131 ***	−0.009	−0.010	−0.088 ***	−0.111 ***
	(0.007)	(0.007)	(0.009)	(0.014)	(0.015)
Negative B/M	0.358 ***	−0.015	−0.051	0.293 **	0.342 ***
	(0.102)	(0.026)	(0.034)	(0.136)	(0.113)
Momentum	−0.014 **	−0.024 **	−0.023	−0.051 ***	−0.051 ***
	(0.007)	(0.012)	(0.015)	(0.013)	(0.011)
Idiosyncratic Risk	0.289	1.123	2.151 **	−4.053 ***	−2.542 ***
	(0.285)	(0.697)	(0.910)	(0.627)	(0.547)
Four-Factor Loadings	Yes	Yes	Yes	Yes	Yes
Industry Dummies	No	No	No	No	No
Firm Fixed Effect	Yes	Yes	Yes	Yes	Yes
Weak Instrument Test:
Cragg–Donald’s Wald F-Statistic	90.7			48.1	46.3
Stock–Yogo Critical Value	22.30			21.28	21.28
Overidentification Test:
Hansen J (p-value)		0.42	0.46	0.36	0.32
Panel B: Two-Stage Least Square Test for COVID-19 Period
Variables	1st Stage	2nd Stage		Lewbel
	(1)	(2)	(3)	(4)	(3)
	KC	Raw Returns	Abn. Return	Raw Returns	Abn. Return
KC		0.311 **	0.362 ***	0.013	0.017
		(0.124)	(0.126)	(0.049)	(0.041)
Crisis		−0.079 ***	−0.076 ***	0.010	0.007
		(0.028)	(0.029)	(0.008)	(0.012)
KC×Crisis		0.107 ***	0.112 ***	0.089 ***	0.105 ***
		(0.041)	(0.043)	(0.034)	(0.039)
Ind. Mean KC (Inst.Var.)	0.028 ***
	(0.010)
2 Years Lag KC (Inst. Var.)	0.182 ***
	(0.025)
State Mean KC (Inst. Var.)	0.037 ***
	(0.013)
Market Cap.	−0.005 ***	0.060	0.075 *	−0.047 ***	−0.056 **
	(0.001)	(0.043)	(0.044)	(0.008)	(0.023)
Long-term Debt	0.025 **	0.051	0.054	0.001	−0.122 *
	(0.010)	(0.089)	(0.044)	(0.044)	(0.072)
Short-term Debt	0.036	0.159	0.124	−0.072	−0.078
	(0.038)	(0.162)	(0.1625	(0.053)	(0.150)
Cash Holdings	−0.052 **	−0.124	−0.062	−0.024	0.015
	(0.022)	(0.097)	(0.099)	(0.071)	(0.094)
Profitability	0.006	−0.308 **	−0.317 ***	0.201 **	−0.213 *
	(0.021)	(0.114)	(0.116)	(0.085)	(0.119)
Book-to-Market	−0.004	−0.057 **	−0.086 ***	0.012	−0.111 ***
	(0.003)	(0.023)	(0.024)	(0.016)	(0.015)
Negative B/M	−0.009	0.060	0.102	−0.031	0.342 ***
	(0.010)	(0.214)	(0.218)	(0.032)	(0.113)
Momentum	−0.030 ***	−0.064 ***	−0.063 ***	−0.011	−0.051 ***
	(0.004)	(0.014)	(0.015)	(0.012)	(0.011)
Idiosyncratic Risk	−0.979 ***	−4.638 ***	−2.325 ***	2.891 *	−2.542 ***
	(0.269)	(0.591)	(0.603)	(1.486)	(0.547)
Four-Factor Loadings	Yes	Yes	Yes	Yes	Yes
Industry Dummies	No	No	No	No	No
Firm Fixed Effect	Yes	Yes	Yes	Yes	Yes
Weak Instrument Test:
Cragg–Donald’s Wald F-Statistic	820.1			55.1	46.3
Stock–Yogo Critical Value	22.30			21.01	21.01
Overidentification Test:
Hansen J (p-value)		0.34	0.37	0.34	0.32

* p < 0.10, ** p < 0.05, and *** p < 0.01.

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.

© 2024 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).

Share and Cite

MDPI and ACS Style

Lee, C.C.; Atukeren, E.; Kim, H. Knowledge Capital and Stock Returns during Crises in the Manufacturing Sector: Moderating Role of Market Share, Tobin’s Q, and Cash Holdings. Risks 2024, 12, 100. https://doi.org/10.3390/risks12060100

AMA Style

Lee CC, Atukeren E, Kim H. Knowledge Capital and Stock Returns during Crises in the Manufacturing Sector: Moderating Role of Market Share, Tobin’s Q, and Cash Holdings. Risks. 2024; 12(6):100. https://doi.org/10.3390/risks12060100

Chicago/Turabian Style

Lee, Chaeho Chase, Erdal Atukeren, and Hohyun Kim. 2024. "Knowledge Capital and Stock Returns during Crises in the Manufacturing Sector: Moderating Role of Market Share, Tobin’s Q, and Cash Holdings" Risks 12, no. 6: 100. https://doi.org/10.3390/risks12060100

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Menu

Knowledge Capital and Stock Returns during Crises in the Manufacturing Sector: Moderating Role of Market Share, Tobin’s Q, and Cash Holdings

Abstract

1. Introduction

2. Data and Methods

2.1. Sample and Data

2.2. Variables

2.2.1. Dependent Variable; Stock Returns

2.2.2. Main Independent Variable; Knowledge Capital

2.2.3. Moderators; Market Share, Tobin’s Q, and Cash Holdings

2.3. Empirical Model

2.4. Summary Statistics

3. Results and Analyses

3.1. KC and Crisis Returns during GFC

3.2. KC and Crisis Returns during COVID-19

3.3. Moderating Role of Market Share, Tobin’s Q and Cash Holdings

3.4. Robustness

4. Conclusions

Author Contributions

Funding

Data Availability Statement

Conflicts of Interest

Appendix A. Variable Definitions and Sources

Appendix B. Descriptive Statistics for Non-Crisis and Crisis Periods

Appendix C. State-Level Cluster–Robust Standard Error Test

References

Share and Cite

Article Metrics

Article Access Statistics

Further Information

Guidelines

MDPI Initiatives

Follow MDPI