Climate Policy Uncertainty and Corporate Investment: Evidence From the U.S. Tourism and Hospitality Sector

Abstract

We examine the association between climate policy uncertainty (CPU) and corporate investment using U.S. data on tourism and hospitality firms over the period 2001–2020. We find that the aggregate country-level CPU is significantly negatively associated with capital investment, with the impact extending at least 4 years into the future. In particular, we find that CPU has an influence on firms’ capital investment that is incremental to economic policy uncertainty (EPU). By contrast, we document that CPU is unrelated to R&D investment. Our main findings are robust to a battery of sensitivity tests, including an instrumental variable approach and sub-industry analyses. Overall, our findings highlighting the impact of CPU on capital investment have significant implications for academics, managers, and policy makers.

Keywords

climate policy uncertainty tourism capital investment R&D investment economic policy uncertainty

Introduction

Climate change has been acknowledged as a grand environmental and social challenge around the world. According to the recent reports released by World Economic Forum (WEF, 2020, 2021), climate risk has been continuously ranked as one of the top risks facing firms around the globe in terms of likelihood and impact.¹ Climate change brings about significant uncertainty arising from both physical risks and transition risks (e.g., Ernst&Young, 2016; Standard & Poor’s, 2017). In combating these risks, government plays an important role in enacting climate-related policies to cope with climate change (e.g., Becken & Loehr, 2022; Glass& Newig, 2019). However, there is a significant amount of uncertainty regarding the exact timing and economic consequence of these policies (e.g., Bartram et al., 2022). For example, Bartram et al. (2022) show that climate policy may create unintended consequences by using the California cap-and-trade program as an example. Therefore, it is warranted to examine how uncertainty arising from climate policy affects corporate investment decisions.

The purpose of the present study is to examine the impact of CPU on corporate investment decisions of U.S. tourism and hospitality firms. We focus on the U.S. tourism and hospitality sector because it is a major contributor to greenhouse gas emissions as well as the U.S. economy in terms of economic output and employment (World Travel and Tourism Council [WTTC], 2021).² In addition, the U.S. has long been recognized as one of the top tourist destinations because of the quality of its tourism goods and services. Further, focusing exclusively on the tourism and hospitality firms allows us to account for potential confounding factors in a way that is generally impossible in a cross-industry analysis.

Prior literature documents various impacts exerted by climate change on the tourism sector (e.g., Dillimono & Dickinson, 2015; Kaján & Saarinen, 2013). A recent strand of literature examines the impact of various uncertainties arising from economic policy (Demir & Gözgör, 2018), geopolitical risk (Demir et al., 2019), and country risk (C. C. Lee & Chen, 2021) on the tourism and hospitality industries. In particular, a growing strand of literature investigates whether and how economic policy uncertainty (EPU) influences the tourism industry in terms of corporate investment (Akron et al., 2020), inbound tourist arrivals (Khan et al., 2021), and tourism consumption (Nguyen et al., 2022). As a highly climate-sensitive industry, tourism is likely to be affected by climate change (e.g., Scott, 2011; Steiger et al., 2019). However, there is a dearth of research exploring the impact of CPU on the tourism and hospitality sector. We, therefore, aim to fill this gap in the literature.

Drawing on real options theory, we argue that climate policy uncertainty is negatively associated with both capital and R&D investment. Real options theory predicts that firms postpone their investment until uncertainty is resolved (e.g., Dixit & Pindyck, 1994). Consistent with this notion, prior literature documents the negative impact of various uncertainties on corporate investment (e.g., Gulen & Ion, 2016; Kanagaretnam et al., 2022). However, existing literature documents inconsistent evidence on the relationship between uncertainty and corporate investment (e.g., Abel, 1983; Gulen & Ion, 2016; Hartman, 1972; Segal et al., 2015). For example, earlier studies document a positive association between uncertainty and corporate investment (e.g., Abel, 1983; Hartman, 1972). Furthermore, a recent stand of literature finds that corporate investment is a potential means to adapt to climate change (Barreca et al., 2015, 2016), suggesting a positive relationship between climate change and corporate investment. Taken together, whether and to what extent climate policy uncertainty is related to tourism and hospitality firms’ investment is ultimately an empirical question.

One of the main challenges in examining the impact of climate policy uncertainty on corporate investment is to find an appropriate measure of CPU. Fortunately, our measure of CPU is obtained from Gavriilidis (2021), who follows the approach proposed by Baker et al. (2016) building the EPU index and constructs the monthly CPU index. Specifically, the CPU index is constructed by conducting textual analysis using three groups of keywords related to climate change, uncertainty, and regulation, respectively, in eight leading U.S. newspapers.³

Using the CPU index proposed by Gavriilidis (2021) over the period 2001–2020, we find that CPU is negatively related to capital investment but unrelated to R&D investment for the U.S. tourism and hospitality firms. In terms of economic magnitude, a one standard deviation increase in CPU is associated with a 15.7% decrease in capital investment. In particular, we control for EPU to ensure that the effect we are examining cannot be attributed to EPU, a source of uncertainty influencing corporate investment as identified in the prior literature (e.g., Gulen & Ion, 2016; Kanagaretnam et al., 2022). We find that the effect of CPU on capital investment is incremental to EPU, highlighting the important role of CPU in influencing tourism and hospitality firms’ investment decisions.

From a long-horizon perspective, we show that the effect of climate policy uncertainty on capital investment extends at least 4 years into the future, highlighting the long-lasting impact of climate policy uncertainty. Further analyses indicate that our main findings continue to hold at a sub-industry level. Our main findings are robust to an instrumental variable analysis and an additional battery of robustness tests. Finally, we document that our main results continue to hold even after controlling for uncertainty arising from the pandemic COVID-19.

We contribute to the literature in two important ways. First, we contribute to the literature examining the impact of climate change on tourism by focusing on the impact of CPU on corporate investment for tourism and hospitality firms. Prior literature has examined the impact of CPU on price dynamics of stocks (Bouri et al., 2022), and stock returns (Chan & Malik, 2022). We add to this literature by examining its impact on corporate investment. Specifically, unlike existing literature that focuses largely on aggregated corporate investment (e.g., Gulen & Ion, 2016), we examine the differential impact of CPU on capital and R&D investment, respectively.

Second, we extend the literature that examines the impact of policy uncertainty on corporate behaviors for tourism firms (Akron et al., 2020; Demir et al., 2019; Hoang, 2022). Unlike previous literature that focuses extensively on EPU, we focus on the impact of CPU on corporate investment. To the best of our knowledge, this is the first empirical study examining the impact of CPU on investment decisions of tourism and hospitality firms. Specifically, we complement Hoang (2022) by showing that while climate policy uncertainty is negatively related to capital investment, its influence on R&D investment of tourism and hospitality firms is negligible. In particular, we show the incremental role of climate change uncertainty beyond economic policy uncertainty, highlighting the role of CPU in shaping capital investment.

The rest of the study is organized as follows. Section 2 discusses relevant literature and hypothesis. Section 3 presents data sources, variable definitions, and empirical methodology. Section 4 presents the empirical results. Section 5 discusses relevant policy implications, and the final section concludes the study.

Relevant Literature and Hypotheses Development

Related Literature

The relationship between tourism and climate change has been studied extensively in the literature (see, e.g., Arabadzhyan et al., 2021; Kaján and Saarinen, 2013; Steiger et al., 2019, for a review). For example, Kaján and Saarinen (2013) systematically review the academic literature on tourism and climate change, with an emphasis on adaptation prior to 2012. Subsequent studies are conducted either at a sub-industry level (e.g., Steiger et al., 2019) or with a different emphasis (e.g., Arabadzhyan et al., 2021). While earlier literature focuses primarily on how climatic conditions influence tourism demand (e.g., Eugenio-Martin & Campos-Soria, 2010; Martín, 2005), recent literature typically involves developing mitigation and adaptation strategy in the tourism domain (e.g., Dogru et al., 2019; Schliephack & Dickinson, 2017; Scott et al., 2012). Our study extends the latter stream of research by focusing on the adaptation strategy of tourism and hospitality firms in terms of corporate investment.

Our study is related to a growing strand of literature examining the influence of uncertainty on corporate investment (Gulen & Ion, 2016; Jens, 2017; Kanagaretnam et al., 2022). For example, using a difference-in-differences research design, Jens (2017) shows that political uncertainty is negatively associated with firms’ investment. Specifically, the decline in investment is about 5% before elections and up to 15% for firms sensitive to political uncertainty. Similarly, Gulen and Ion (2016) indicate that the news-based economic policy uncertainty is negatively associated with firm-level capital investment.

Another related strand of literature documents the impact of uncertainty on tourism activities (Akadiri et al., 2020; Akron et al., 2020; Khan et al., 2021). For example, Khan et al. (2021) find that EPU is negatively associated with inbound tourism in the UK. In contrast, Akadiri et al. (2020) examine and document that tourism predicts EPU in a panel of 12 countries in three different continents. Akron et al. (2020) document that EPU is negatively associated with corporate investment in the U.S. tourism sector. In particular, they show that the negative effect is mainly driven by firms with lower quantile of capital expenditure ratio.

Despite a plethora of studies examining the impact of different types of uncertainty and risk on corporate investment, there is a dearth of studies focusing on the influence of policy uncertainty arising from climate change, with a few exceptions (e.g., Bouri et al., 2022; Chan & Malik, 2022; Hoang, 2022). For example, Bouri et al. (2022) find that climate policy uncertainty influences investors’ preference between green and brown energy stocks.

Hypotheses Development

We draw on real options theory (ROT) to investigate the relationship between climate policy uncertainty and corporate investment because our research setting, involving both uncertainty and irreversibility, provides an ideal context for the application of ROT. As suggested by Dixit and Pindyck (1994), the real options approach to making investment indicates that firms may reap benefits if they delay their investment until the ambiguity is mitigated. According to ROT, if a firm can postpone investment, it will make the investment only if the net present value (NPV) of the investment is greater than the value of the option to defer. There is a voluminous literature drawing on ROT to investigate how various uncertainties influence corporate investment (e.g., Gulen & Ion, 2016; Wilson, 2021). For example, Gulen and Ion (2016) find a negative relationship between firm-level investment and aggregated uncertainty associated with future policy and regulatory outcomes.

As suggested by prior literature, climate risk mainly consists of physical risk and transition risk (e.g., Ernst & Young, 2016;Standard & Poor’s, 2017),⁴ both of which could potentially influence CPU, which, in turn, impacts tourism and hospitality firms, especially considering that tourism is a climate-dependent industry. Drawing on real options theory, we argue that uncertainty associated with these risks could defer corporate investment until it is resolved. Thus, managers may take a “wait and see” approach and postpone their investment. Based on these discussions, we propose our first hypothesis as follows (in alternate form):

H1: CPU is negatively associated with capital investment for U.S. tourism and hospitality firms.

We next focus on the impact of CPU on R&D investment because prior literature has highlighted the role of R&D investment in sustaining firms’ growth (Romer, 1990). Managers are less likely to cut R&D investment because doing so may decrease firm value in the long run (Bushee, 1998). R&D investment is also critical to tourism and hospitality firms in that it sustains their long-run performances (e.g., Albaladejo & Martínez-García, 2015; Orfila-Sintes et al., 2005). For example, Albaladejo and Martínez-García (2015) suggest that income from tourism increases with innovation investment by constructing an R&D-based endogenous growth model. In addition, it is worth noting that R&D investment in the tourism industry is a potential adaptation means to address challenges brought about by climate change.

It is widely acknowledged that R&D investment is risky and various aspects of R&D investment can be affected by uncertainty (e.g., Banerjee & Siebert, 2017; Kim & Wilemon, 2002). Thus, managers may be less likely to engage in R&D investment when CPU is high and take a “wait and see” approach to minimize potential losses accordingly. Together, we propose our second hypothesis as follows (in alternate form):

H2: CPU is negatively associated with R&D investment for U.S. tourism and hospitality firms.

However, it is important to note that there are counterarguments supporting a positive relationship between CPU and capital and R&D investment. Earlier literature has documented that uncertainty is positively associated with corporate investment (e.g., Abel, 1983; Hartman, 1972). For instance, Abel (1983) shows that higher uncertainty is associated with a higher investment rate regardless of the curvature of the marginal adjustment cost function. Extending this logic to our context, it is plausible that CPU can be positively associated with corporate investment as well. In a similar vein, Barreca et al. (2015, 2016) find that potential climate physical damage decreases with both capital and technological innovation investment. Thus, firms may increase corporate investment as an adaptation strategy against climate change. Following this line of reasoning, we expect a positive relationship between CPU and corporate investment. Taken together, whether and how CPU is related to corporate investment is ultimately an empirical question.

Empirical Design

Sample Selection

Our data are collected from several sources. Specifically, we obtain U.S. climate policy uncertainty data from Gavriilidis (2021). We collect financial data from Compustat for tourism and hospitality firms in the following four sub-industries, namely airline, hotel, restaurant, and casino. We obtain U.S. GDP data from Federal Reserve Bank of St. Louis and consumer confidence index data from OECD (2022). We exclude firms with missing information on all variables used in equation (1). The intersection of these data sets leads to a final sample of 2,957 firm-year observations, representing 307 distinct tourism and hospitality firms over the period 2001–2020.

The CPU Index

The CPU index has been employed in recent economics and finance research (e.g., Bouri et al., 2022; Chan & Malik, 2022; Hoang, 2022). Following the methodology in Baker et al. (2016), Gavriilidis (2021) constructs the CPU index. Gavriilidis (2021) began to search for articles in eight leading U.S. newspapers containing three groups of keywords related to uncertainty, climate change, and regulation, respectively, from 2000 to 2021. Then, Gavriilidis (2021) standardized each series from eight newspapers with a standard deviation of one and averaged across newspapers by month. Finally, the averaged series are normalized to have a mean value of 100 over the period 2000–2021 by month. Following prior literature, we transform the monthly index into yearly data by using the weighted average method and measure CPU by taking the natural logarithm of the annual averaged series.

Figure 1 graphically presents the CPU index. Similar to the EPU index developed by Baker et al. (2016), it demonstrates significant time-series variation. A close visual inspection of the index suggests that it spikes during periods characterized by higher uncertainty associated with climate-induced extreme weather events. As shown in Figure 1, for example, there is a spike around 2005, coinciding with the occurrence of Hurricane Katrina. Similarly, we can observe another spike around 2017, possibly due to either the occurrence of three monster hurricanes (i.e., Harvey, Irma, and Maria) or the Trump administration’s withdrawal from the Paris Climate Change Agreement in 2017.

Measures of Corporate Investment

We proxy for corporate investment using two different measures: capital investment (Capx) and R&D investment (R&D). Capx is measured as the ratio between capital investment and lagged total assets, while R&D is measured as the ratio between R&D investment and lagged total assets. Following prior literature, R&D is set to zero if it is missing in Compustat. Given that prior literature has highlighted the role of R&D investment in the tourism sector (e.g., Albaladejo & Martínez-García, 2015; Orfila-Sintes et al., 2005; Vu & Hartley, 2022), unlike Gulen and Ion (2016) who focus exclusively on capital expenditure, we take R&D investment into account to present a more complete picture on whether and how CPU affects corporate investment.

Control Variables

Following Gulen and Ion (2016), we control for firms’ cash flow (CF), Tobin’s q (TQ), and sales growth (Saleg). CF is measured as the ratio between cash flows and lagged total assets. Kaplan and Zingales (1997) suggest that the coefficient on CF measures investment-cash flow sensitivity. TQ is measured as the sum of the market value of equity plus the book value of assets minus book value of equity plus deferred taxes, divided by lagged total assets. Saleg is defined as the year-on-year changes in sales. Both TQ and Saleg are widely used to capture investment opportunities (e.g., Gulen & Ion, 2016; Julio & Yook, 2012). Finally, following Gulen and Ion (2016), we control for forecasted GDP growth (GDPg) and consumer confidence index (CCI) that represent investment opportunities. We winsorize all variables at the 1st and 99th percentiles to mitigate the influence of potential extreme outliers.

Empirical Model

We build on prior literature (e.g., Gulen & Ion, 2016; Kanagaretnam et al., 2022) and specify our regression model as follows:⁵

\begin{array}{l} I n v e s t m e n t_{i t + 1} = α_{i} + β_{1} C P U_{t} + β_{2} C F_{i t} \\ + β_{3} T Q_{i t} + β_{4} S a l e g_{i t} + β_{5} G D P g_{t} \\ + β_{6} C C I_{t} + F i r m F E_{i} + ε_{i t} \end{array}

(1)

where i and t denote firm and year, respectively. Investment denotes either capital or R&D investment. The key variable of interest is CPU, measured as the natural logarithm of CPU index averaged at the yearly level, as aforementioned. CF, TQ, and Saleg denote cash flows, Tobin’s q, and sales growth, respectively, as previously defined. Firm FE denotes firm fixed effects that are used to control for time-invariant firm-level characteristics. Following Gulen and Ion (2016), we don’t include year fixed effects because it is likely to subsume the effect of CPU. Instead, we control for forecasted GDP growth (GDPg) and consumer confidence index (CCI), both of which represent investment opportunities. ε_it is the error term. We cluster standard errors at the firm level (Petersen, 2009). The coefficient of interest is β₁. We expect the sign of β₁ to be negative if CPU dampens corporate investment (i.e., capital and R&D investment) in the tourism and hospitality industry.

Descriptive Statistics

Table 1 reports the sample distribution of firms based on sub-industry classification. The hotel sector constitutes the largest proportion of observations in our sample (42.3%), followed by the casino sector (24.7%), the airline sector (19.5%), and the restaurant sector (13.5%). We observe no significant sector bias in our sample distribution.

Table 1.

Sample Distribution by Sub-Industry.

	Airline	Hotel	Restaurant	Casino	Total
Observations	576	1,251	401	729	2,957
Firms	50	138	45	74	307

Table 2 provides descriptive statistics for the main variables in equation (1). The means (medians) of capital and R&D investment divided by lagged total assets are 0.093 (0.066) and 0.001 (0), respectively. The standard deviations of Capx and R&D are 0.095 and 0.009, respectively, suggesting substantial variation in both capital and R&D investment across firms. The mean (median) of CPU is 4.329 (4.436). The standard deviation of CPU is 0.58, implying significant time-series variation. The descriptive statistics of other variables are largely consistent with prior research.

Table 2.

Descriptive Statistics.

	N	Mean	Std. dev.	p25	Median	p75
Capx	2,957	0.093	0.095	0.029	0.066	0.124
R&D	2,957	0.001	0.009	0	0	0
CPU	2,957	4.329	0.58	4.083	4.436	4.781
CF	2,957	0.095	0.172	0.048	0.099	0.163
TQ	2,957	1.98	3.597	1.068	1.373	1.963
Saleg	2,957	0.124	0.435	-0.017	0.057	0.148
GDPg	2,957	0.039	0.02	0.03	0.04	0.049
CCI	2,957	4.603	0.013	4.596	4.607	4.612

Table 3 presents Pearson pairwise correlations among the main variables used in equation (1). As shown in Table 3, we find that CPU has a positive correlation with Capx at the 1% level and R&D at the 5% level. Both TQ and Saleg are strongly positively correlated with Capx and R&D, while CF is strongly positively correlated with Capx but strongly negatively correlated with R&D. However, it is worthwhile to note that these correlation coefficients merely provide some preliminary relationships because they are based on univariate analyses. We thus postpone our inferences to the multivariate analyses discussed in the next section.

Table 3.

Pearson Pairwise Correlations.

Variables	Capx	R&D	CPU	CF	TQ	Saleg	GDPg	CCI
Capx	1.000
R&D	−.074***	1.000
CPU	−.078***	−.043**	1.000
CF	.195***	−.167***	.035*	1.000
TQ	.060***	.072***	−.042**	−.451***	1.000
Saleg	.160***	.046**	−.028	.043**	−.028	1.000
Gdpg	.018	.009	−.537***	.022	.066***	.029	1.000
CCI	.055***	.003	−.167***	.009	.073***	.013	.490***	1.000

***

Significance at 1%.

Significance at 5%.

Significance at 10%.

Empirical Results

Main Results

We present our main regression results on the relationship between climate policy uncertainty and corporate investment in Table 4. Specifically, Table 4 Column (1) reports the relationship between CPU and capital investment, while Column (2) of Table 4 presents the association between CPU and R&D investment.

Table 4.

Main OLS Regression Results.

	(1)	(2)
	Capx	R&D
CPU	−0.0251*** (−5.93)	−0.0002 (−1.05)
CF	0.0409 (1.26)	−0.0029** (−2.22)
TQ	0.0040*** (3.36)	−0.0001 (−0.70)
Saleg	0.0119** (2.40)	−0.0001 (−0.28)
GDPg	0.1154 (1.32)	0.0019 (0.56)
CCI	0.4027** (2.56)	−0.0025 (−0.52)
Constant	−1.6774** (−2.34)	0.0137 (0.62)
Firm FE	Yes	Yes
N	2,957	2,957
Adj. R²	.3844	.8682

Note. This table presents the regression results on the relation between CPU and corporate investment. The t-statistics (reported in parentheses) are based on standard errors clustered at the firm level.

***

Significance at 1%.

Significance at 5%.

Significance at 10%.

We find that CPU is negatively associated with capital investment. Specifically, the coefficient on CPU (coef. = −0.0251, t-stat. = −5.93) is negative and statistically significant at the 1% level in our main regression model, consistent with H1. In terms of economic magnitude, a one standard deviation increase in CPU is associated with a 15.7% decrease in capital investment.⁶ The signs and magnitudes of the control variables are largely consistent with prior literature (e.g., Gulen & Ion, 2016). For example, we find that both Tobin’s q and sales growth are positively associated with capital investment for U.S. tourism and hospitality firms, consistent with Gulen and Ion (2016). Nevertheless, the effect of cash flows on capital investment is positive but insignificant.

By contrast, we fail to document a significant relationship between CPU and R&D investment in Column (2) ofTable 4. The coefficient on CPU (coef. = −0.0002, t-stat. = −1.05) is negative but insignificant at the 5% level, thereby rejecting H2. The fact that failing to find a negative relationship between CPU and R&D investment suggests that managers are less likely to defer R&D investment, in part because it plays an increasingly important role in climate adaptation.

The Long-Term Effect of CPU on Corporate Investment

Having established the short-term effect of CPU on capital investment, we proceed to investigate how the relationship evolves over time because adjustment in investment plan can take a quite long period of time. We report the long-term effect of CPU on corporate investment in Table 5.

Table 5.

Prolonged Period Analyses.

	(1)	(2)	(3)	(4)	(5)	(6)
	T = 2		T = 3		T = 4
	Capx	R&D	Capx	R&D	Capx	R&D
CPU	−0.0301*** (−6.85)	−0.0002 (−0.75)	−0.0387*** (−9.84)	−0.0002 (−0.95)	−0.0330*** (−7.71)	0.0001 (0.42)
CF	0.0755*** (3.03)	−0.0031** (−1.98)	0.0690** (2.20)	−0.0010 (−1.07)	0.0325 (1.24)	0.0010 (0.85)
TQ	0.0069*** (4.54)	−0.0001 (−1.36)	0.0030*** (3.00)	−0.0000 (−1.57)	−0.0007 (−0.54)	0.0000 (0.30)
Saleg	0.0026 (0.52)	−0.0004 (−1.04)	0.0011 (0.20)	−0.0003* (−1.71)	0.0013 (0.32)	0.0005 (1.16)
GDPg	0.2599*** (2.86)	−0.0009 (−0.26)	−0.0964 (−0.99)	−0.0041 (−0.90)	−0.2870*** (−2.88)	0.0042 (0.69)
CCI	−0.0922 (−0.58)	0.0021 (0.48)	−0.0863 (−0.48)	0.0039 (0.73)	−0.3386* (−1.80)	0.0000 (0.00)
Constant	0.6059 (0.83)	−0.0071 (−0.36)	0.6362 (0.77)	−0.0156 (−0.65)	1.7881** (2.07)	0.0003 (0.01)
Firm FE	Yes	Yes	Yes	Yes	Yes	Yes
N	2,624	2,624	2,323	2,323	2,052	2,052
Adj. R²	.4121	.8500	.3894	.8794	.3906	.8687

Note. This table presents the regression results on the relation between CPU and corporate investment into the future periods. The t-statistics (reported in parentheses) are based on standard errors clustered at the firm level.

***

Significance at 1%.

Significance at 5%.

Significance at 10%.

As shown in Table 5, we find the coefficients on CPU are negative and statistically significant in Columns (1), (3), and (5), suggesting that the impact of CPU on capital investment extends at least 4 years into the future. Specifically, the impact peaks in Year 3 and begins to decay afterwards (but still negative and significant). These findings provide additional corroborating evidence to our main findings and highlight the long-lasting impact of CPU. In addition, we find the coefficients on CPU are small and insignificant in Columns (2), (4), and (6), indicating that CPU and R&D investment remain unrelated in prolonged periods, consistent with our main findings.

Sub-Industry Analyses

We proceed to examine whether our main findings continue to hold for each of the four sub-industries, namely airline, hotel, restaurant, and casino.⁷ The rationale behind this analysis is that these four sub-industries are unlikely to be equally exposed to climate policy uncertainty. Consistent with this view, prior literature shows that treating these sub-industries homogenously may account for the inconsistent findings on the tourism-economic growth nexus (Tang & Jang, 2009). As a result, a separate analysis at the sub-industry level can provide a deeper understanding of the relationship between climate policy uncertainty and corporate investment, which has significant implications for academics and managers.

We report sub-industry estimation results in Table 6. We consistently find that CPU exhibits a negative and significant relationship with capital investment across all four sub-industries. Compared with the main results in Table 4, the magnitude of the impact of CPU on corporate investment is stronger in airline and hotel sub-industries but weaker (still significant) in restaurant and casino sub-industries. One possible explanation is that both aviation and hotel sub-industries are more energy-intensive relative to the remaining sub-industries, thereby rendering firms in these sub-industries more susceptible to climate policy uncertainty. Consistent with our main findings, we fail to document a negative relationship between CPU and R&D investment. Overall, our sub-industry analyses lend credence to our main findings.

Table 6.

Sub-Industry Analyses.

	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)
	Airline		Hotel		Restaurant		Casino
	Capx	R&D	Capx	R&D	Capx	R&D	Capx	R&D
CPU	−0.0326*** (−2.81)	−0.0003 (−0.80)	−0.0293*** (−4.91)	0.0001 (0.67)	−0.0115** (−2.16)	−0.0000 (−1.02)	−0.0184* (−1.92)	−0.0005 (−0.96)
CF	0.0348 (0.34)	−0.0058 (−1.11)	0.0450 (0.85)	−0.0005 (−0.53)	0.1487 (1.36)	0.0000 (0.92)	−0.0128 (−0.36)	−0.0018 (−1.38)
TQ	0.0034 (1.18)	−0.0000 (−0.25)	0.0031* (1.72)	−0.0002 (−1.53)	0.0080 (0.68)	−0.0000 (−0.59)	0.0114*** (3.13)	−0.0000 (−0.04)
Saleg	0.0083 (0.48)	0.0013 (1.65)	0.0203 (1.54)	0.0000 (0.01)	−0.0043 (−0.52)	−0.0000 (−0.95)	0.0104* (1.72)	−0.0003 (−0.65)
GDPg	−0.1683 (−0.55)	−0.0068 (−0.67)	0.1772 (1.28)	0.0011 (0.22)	0.2132 (1.50)	−0.0000 (−0.97)	0.1862 (1.38)	0.0081 (1.29)
CCI	0.8089* (1.98)	0.0001 (0.01)	0.1883 (0.70)	0.0024 (0.48)	−0.1205 (−0.36)	0.0002 (1.01)	0.5984*** (2.68)	−0.0100 (−1.05)
Constant	−3.4860* (−1.89)	0.0030 (0.06)	−0.6594 (−0.53)	−0.0106 (−0.47)	0.6240 (0.40)	−0.0011 (−1.00)	−2.6401** (−2.56)	0.0513 (1.15)
Firm FE	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
N	576	576	1,251	1,251	401	401	729	729
Adj. R²	.3355	.8204	.4111	.4035	.4093	.7738	.2505	.9222

Note. This table presents the regression results of the relation between CPU and corporate investment for four sub-industries. The t-statistics (reported in parentheses) are based on standard errors clustered at the firm level.

***

Significance at 1%.

Significance at 5%.

Significance at 10%.

Omitted Variable—The Incremental Effect of CPU

A potential challenge of our study is to ensure that we are capturing the effect of CPU rather than the effect of other sources of uncertainty, because prior literature (e.g., Akron et al., 2020; Gulen & Ion, 2016) documents that EPU dampens corporate investment. In this section, we examine whether CPU has an incremental effect on corporate investment beyond EPU. To this end, we collect EPU data from Baker et al. (2016) and augment equation (1) by adding the variable of EPU.

Table 7 presents the estimation results of equation (1) after controlling for EPU. Consistent with prior literature (Akron et al., 2020; Gulen & Ion, 2016), we document a negative relationship between EPU and capital investment. Specifically, the coefficient on EPU (coef. = −0.0313, t-stat. = −4.26) is negative and statistically significant at the 1% level. More importantly, we continue to document a negative relationship between CPU and capital investment and no association between CPU and R&D investment even after including EPU as an additional control variable. The coefficient on CPU (coef. = −0.0221, t-stat. = −5.11) is negative and statistically significant at the 1% level, suggesting the incremental effect of CPU on capital investment beyond EPU. In particular, we find that the economic magnitude of the effect of CPU is similar to that of EPU, highlighting that tourism and hospitality firms need to pay attention to both types of uncertainty when making capital investment decisions.

Table 7.

The Incremental Impact of CPU.

	(1)	(2)
	Capx	R&D
CPU	−0.0221*** (−5.11)	−0.0002 (−0.94)
EPU	−0.0313*** (−4.26)	−0.0001 (−0.34)
CF	0.0396 (1.22)	−0.0029** (−2.22)
TQ	0.0039*** (3.34)	−0.0001 (−0.70)
Saleg	0.0102** (2.02)	−0.0001 (−0.31)
GDPg	−0.0026 (−0.03)	0.0014 (0.49)
CCI	0.3708** (2.34)	−0.0026 (−0.52)
Constant	−1.3889* (−1.90)	0.0150 (0.61)
Firm FE	Yes	Yes
N	2,957	2,957
Adj. R²	.3926	.8682

Note. This table presents the regression results of the incremental effects of CPU beyond EPU. The t-statistics (reported in parentheses) are based on standard errors clustered at the firm level.

***

Significance at 1%.

Significance at 5%.

Significance at 10%.

Instrumental Variable Analysis

Given that climate policy uncertainty might not be a strictly exogenous variable, we further mitigate the endogeneity concern by performing an instrumental variable analysis. An appropriate instrument should be highly correlated with CPU and affects corporate investment only through this association. Given that prior literature (e.g., McCright & Dunlap, 2011) has highlighted the impact of political ideology on policies regarding climate change, following Gulen and Ion (2016), we employ the first dimension of the DW-NOMINATE scores for the U.S. senate as an instrument for CPU. McCarty et al. (1997) construct these scores to track the legislators’ ideological positions over time. Our instrumental variable is calculated as the average of scores of Republican senators minus the average for Democratic senators.

In line with Gulen and Ion (2016), we perform a time-series regression in the first stage, where we regress CPU on the DW-NOMINATE score and the industry-wide average of cash flows, Tobin’s q, and sales growth. In the second stage, we perform a panel regression using the fitted value of CPU from the first stage.

We report the instrumental variable analysis results in Table 8. In the first stage, we find the coefficient on DW_NOMINATE (coef. = 5.0596, t-stat. = 31.11) is positive and statistically significant at the 1% level, satisfying the relevance restriction. The Kleibergen-Papp Wald F statistic is greater than the critical value of the Stock and Yogo (2005) statistic, implying that DW_NOMINATE is not a weak instrument. The exclusion restriction is also satisfied because we are unaware of any channels DW_NOMINATE can affect corporate investment directly other than through affecting climate policy uncertainty. Therefore, the exclusion condition is satisfied.⁸ In the second stage, we find that the coefficient on Pred_CPU is negative and statistically significant at the 1% level in Column (2) (coef. = −0.0561, t-stat. = −8.62) but insignificant in Column (3) (coef. = −0.0002, t-stat. = 0.74), consistent with our main findings. Overall, the results from the instrumental variable approach provide corroborating evidence to our main findings.

Table 8.

Instrumental Variable Analyses.

	(1)	(2)	(3)
		Second stage
	First stage	Capx	R&D
DW_NOMINATE	5.0596*** (31.11)
Pred_CPU		−0.0561*** (−8.62)	−0.0002 (−0.74)
CF	−0.0228 (−0.35)	0.0405* (1.72)	−0.0029** (−2.02)
TQ	−0.0018 (−0.78)	0.0040** (2.37)	−0.0001 (−0.66)
Saleg	0.0018 (0.09)	0.0099** (2.35)	−0.0001 (−0.26)
GDPg	−11.0387*** (−34.10)	−0.2694** (−2.45)	0.0010 (0.29)
CCI	−4.4176*** (−6.88)	0.5215*** (4.66)	−0.0022 (−0.62)
Constant	21.4117*** (7.37)	−2.1090*** (−4.16)	0.0114 (0.69)
Kleibergen-Papp Wald F	967.80
Firm FE	Yes	Yes	Yes
N	2,957	2,957	2,957
Adj. R²	.6121	.4344	.8821

Note. This table presents the regression results on the relation between CPU and corporate investment using the instrumental variable approach. The t-statistics (reported in parentheses) are based on standard errors clustered at the firm level.

***

Significance at 1%.

Significance at 5%.

Significance at 10%.

Alternative Measure of Climate Change Uncertainty

One concern with our country-level measure of CPU is that although it demonstrates substantial time series variation, it lacks cross-sectional variation. It is plausible that even in the same industry firms may face different levels of uncertainty given that they are exposed to climate change differentially (Sautner et al., 2020). To this end, we use a firm-level climate risk index (CRI) from Sautner et al. (2020) who build the index using earning conference call transcripts between 2002 and 2019 and investigate how uncertainty stemming from regulatory shocks influences corporate investment.

Since we are no longer interested in the impact of CPU on corporate investment, following Gulen and Ion (2016), we replace macroeconomic factors with year fixed effects. We report the results based on an alternative measure of CPU in Table 9. We find that the coefficient on CRI (coef. = −0.0481, t-stat. = −2.26; coef. = −0.0382, t-stat. = −1.34) is negative and statistically significant at the 5% level in Column (1) but insignificant in Column (2), consistent with our main findings in Table 4. Overall, these findings based on a firm-level measure lend further support to our main findings.

Table 9.

An Alternative Measure of CPU.

	(1)	(2)
	Capx	R&D
CRI	−0.0481** (−2.26)	−0.0382 (−1.34)
CF	0.1647*** (2.91)	−0.0007 (−0.73)
TQ	0.0173*** (3.59)	−0.0001 (−0.29)
Saleg	−0.0026 (−0.44)	0.0005 (1.24)
Constant	0.0385*** (4.35)	0.0013*** (7.43)
Firm FE	Yes	Yes
Year FE	Yes	Yes
N	1,349	1,349
Adj. R²	.5287	.9594

Note. This table presents the regression results of the relation between CPU and corporate investment based on an alternative measure of CPU. The t-statistics (reported in parentheses) are based on standard errors clustered at the firm level.

***

Significance at 1%.

Significance at 5%.

Significance at 10%.

Additional Robustness Tests

In this section we perform additional tests to further validate our main findings.

Since our sample period includes the global financial crisis, it is plausible that our results may be driven by the 2008–2009 global financial crisis. To mitigate this concern, we drop firm-year observations in these 2 years and re-estimate our regressions. The results reported in Columns (1) and (2) of Table 10 suggest that our main findings hold during non-financial crisis periods, implying that our results are unlikely to be driven by the global financial crisis.

Table 10.

Additional Robustness Tests.

	(1)	(2)	(3)	(4)	(5)	(6)
	Excluding the financial crisis		Pre-2017		Post-2017
	Capx	R&D	Capx	R&D	Capx	R&D
CPU	−0.0245*** (−5.53)	−0.0002 (−1.08)	−0.0170*** (−3.46)	−0.0001 (−0.46)	−0.0195*** (−3.46)	−0.0000 (−0.78)
CF	0.0375 (1.12)	−0.0031** (−2.24)	0.0374 (1.08)	−0.0021** (−2.36)	0.0334 (0.75)	−0.0001 (−0.18)
TQ	0.0039*** (3.27)	−0.0001 (−0.71)	0.0038*** (3.13)	−0.0000 (−0.54)	0.0109*** (4.13)	0.0000 (0.93)
Saleg	0.0132*** (2.62)	−0.0000 (−0.00)	0.0098** (1.98)	−0.0002 (−0.62)	0.0121 (1.60)	−0.0002 (−1.04)
GDPg	0.1089 (0.91)	0.0006 (0.10)	0.0522 (0.54)	0.0007 (0.21)	1.4550*** (8.81)	0.0033* (1.73)
CCI	0.3511** (1.98)	−0.0033 (−0.79)	0.8462*** (4.55)	0.0019 (0.42)	0.0000 (.)	0.0000 (.)
Constant	−1.4419* (−1.78)	0.0175 (0.92)	−3.7469*** (−4.39)	−0.0067 (−0.33)	0.0778** (2.57)	0.0003 (1.44)
Firm FE	Yes	Yes	Yes	Yes	Yes	Yes
N	2,660	2,660	2,554	2,554	381	381
Adj. R²	.3887	.8519	.3993	.8943	.6336	.9504

Note. This table presents the regression results of the robustness tests on the relation between CPU and corporate investment. The t-statistics (reported in parentheses) are based on standard errors clustered at the firm level.

***

Significance at 1%.

Significance at 5%.

Significance at 10%.

In addition, withdrawal from the Paris Agreement by the Trump administration in 2017 is likely to affect the climate policy uncertainty because majorities in every state support the Paris Agreement (Marlon et al., 2017), which in turn affects corporate investment in the U.S. Therefore, one concern is that our findings are likely to be driven by the withdrawal from the Paris Agreement. To mitigate this concern, we partition our sample into two parts: pre-2017 and post-2017 and investigate the relationship between CPU and corporate investment separately.

As reported in Columns (3) to (6) of Table 10, we find consistent evidence that CPU is negatively related to capital investment but unrelated to R&D investment. Specifically, we find that the magnitude of the effect on capital investment is larger after the withdrawal from the Paris Agreement in 2017. More importantly, the coefficient on CPU (coef. = −0.0170, t-stat. = −3.46) in the pre-2017 is negative and significant at the 1% level. Taken together, the coefficients on CPU in both Columns (3) and (5) are negative and statistically significant, mitigating the concern that our findings are driven by Trump administration’s withdrawal from the Paris Agreement. Taken together, these findings from sub-sample analyses provide further support to our main findings by mitigating the likelihood of alternative explanations.

The Impact of COVID-19

It is worthwhile to note that our sample includes the data of year 2020, when the outbreak of COVID-19 took place in China’s Hubei Province and diffused quickly to the rest of the world, including the U.S. The pandemic has significantly negatively affected the entire U.S. economy (e.g., Goldstein et al., 2021), due to mandatory closures of non-essential business and stay-at-home orders. In particular, the tourism and hospitality sectors were among the hardest hit (S. Lee, 2022). Existing research has documented the relationship between COVID-19 and the tourism and hospitality sectors (e.g., Farzanegan et al., 2021; Selvanathan et al., 2022). For example, Selvanathan et al. (2022) find that the volume of inbound and outbound tourism was positively associated with COVID-19 infected cases. Given the tremendous impact of COVID-19 on the tourism industry, it is therefore reasonable to expect that capital and R&D investment are likely to be affected by the uncertainty arising from the pandemic in the tourism and hospitality sectors. Thus, one potential concern is that whether our main findings are driven by the uncertainty associated with COVID-19. To control for the uncertainty arising from the impact of COVID-19, we obtain the Infectious Disease Equity Market Volatility Tracker data from Baker et al. (2020) and use it to proxy for the uncertainty stemming from infectious diseases (Covid19) during the past two decades, including the COVID-19.

We report our regression results based on the Infectious Disease Equity Market Volatility Tracker data in Table 11. We find that the impact of uncertainty stemming from COVID-19 on both capital and R&D investment is negative but insignificant, much smaller than that of CPU in terms of magnitude. More importantly, we document a significant negative relationship between CPU and capital investment and no evidence on the relationship between CPU and R&D investment even after controlling for the uncertainty arising from COVID-19, lending additional support to our main findings.⁹

Table 11.

The Incremental Impact of Covid-19.

	(1)	(2)
	Capx	R&D
CPU	−0.0261*** (−6.74)	−0.0002 (−1.16)
CF	0.0405 (1.25)	−0.0029** (−2.21)
TQ	0.0040*** (3.36)	−0.0001 (−0.70)
Saleg	0.0119** (2.39)	−0.0001 (−0.28)
GDPg	0.0974 (1.19)	0.0020 (0.59)
CCI	0.4102*** (2.65)	−0.0025 (−0.53)
Covid19	−0.0092 (−1.29)	0.0001 (0.15)
Constant	−1.7017** (−2.40)	0.0139 (0.63)
Firm FE	Yes	Yes
N	2,957	2,957
Adj. R²	.3845	.8681

Note. This table presents the regression results on the relation between CPU and corporate investment when the impact of Covid-19 is taken into account. The t-statistics (reported in parentheses) are based on standard errors clustered at the firm level.

***

Significance at 1%.

Significance at 5%.

Significance at 10%.

Discussions and Theoretical and Practical Implications

Climate change risk has consistently been regarded as one of the top risks facing firms in the next decades (WEF, 2020, 2021). As a climate-sensitive industry, tourism is likely to be affected by uncertainty arising from climate change (Kaján & Saarinen, 2013). The evidence that climate policy uncertainty is negatively related to capital investment but unrelated to R&D investment has significant theoretical and practical implications for different economic agents, including managers, policy makers, and regulators in the U.S. and their counterparts in countries that are highly dependent on tourism.

Theoretically, we draw on real options theory and apply it to a new research setting where we focus on the impact of climate policy uncertainty on the corporate investment behaviors of tourism and hospitality firms. To the best of our knowledge, this is the first empirical study that examines the impact of CPU on investment decisions of tourism and hospitality firms. In particular, we show the incremental role of climate change uncertainty beyond economic policy uncertainty.

A decrease in capital investment would likely lead to a decline in profits. Therefore, when managers take a “wait and see” approach to cope with policy uncertainty related to climate change, they may consider undertaking innovation strategies such as R&D investment. Admittedly, R&D investment is risky and subject to uncertainty (e.g., Banerjee & Siebert, 2017; Kim & Wilemon, 2002). However, it has been identified as a feasible solution to combat climate change since both R&D investment and climate change are long-term phenomena (Blanford, 2009). In this sense, successful R&D investment may help firms compensate losses from reduction in capital investment and adapt to climate change.

Prior literature documents that tourism plays an important role in sustaining economic growth (Aratuo & Etienne, 2019; Brida et al., 2016; Pablo-Romero & Molina, 2013) as well as the (bi)causality between tourism development, economic growth, and emissions (e.g., Zhang & Gao, 2016). Consequently, delays in capital investment of tourism firms may impede the overall growth of the economy. Consistent with this notion, Alam and Paramati (2017) find that a 1% increase in tourism investment is associated with a 0.98% increase in tourism development. To the extent that government policy plays an important role in shaping corporate investment and economic growth, uncertainty regarding which policies will be implemented or are being revised and developed generates a wait-and-see incentive. Therefore, regulators should take measurable action to address the urgent and pressing anthropogenic climate change.¹⁰

Our results also have practical implications for both managers and regulators. Managers should be aware of the potential impact of climate policy uncertainty on corporate investment in addition to climate risk. Government may create monetary or tax incentives to ensure that the normal capital investment is not affected by CPU. For example, government may expand new credit channels to reduce financial frictions in the capital market, especially for firms facing significant financial constraints. Expanded access to finance and tax exemption (reduction) can mitigate financial constraints that in turn dampen capital investment. In addition, firms with sufficient cash reserve are less likely to delay their investment. Thus, firms may consider accumulating cash holdings to deal with the financial friction brought about by climate change, especially for financially constrained ones (Denis & Sibilkov, 2010).

Considering the complex nature of climate risk that is typically characterized as a long-term phenomenon with non-linear impact, climate risk assessment should be conducted by government to better integrate tourism and climate change policies (Becken et al., 2020), especially in right-leaning states.¹¹ Integrating climate change policy with tourism policy achieves not only economic returns but also environmental returns (Paramati et al., 2018). In line with this view, Scott (2011) argues that climate change issues are inseparable from sustainable tourism development. In addition, climate mitigation and adaptation measures should be enacted to mitigate the negative impact of climate change at the national, state, and county level, thus indirectly mitigating the impact of climate policy uncertainty.

Conclusions

Climate change has played an increasingly important role in affecting firms’ behaviors. In this study, we examine whether policy uncertainty arising from climate change influences capital and R&D investment of tourism and hospitality firms in the U.S. Our study is the first in terms of examining the differential impact of CPU on capital and R&D investment in the tourism and hospitality sector.

Our results show that climate policy uncertainty is negatively related to capital investment but unrelated to R&D investment. The negative impact of CPU on capital investment extends at least 4 years into the future. In particular, the effect of CPU is incremental to EPU, another important source of uncertainty identified in the literature. Our results continue to hold in a battery of robustness tests, including sub-industry analyses and an instrumental variable analysis.

Our study contributes to the literature in two main aspects. First, we contribute to the literature examining the impact of climate change on tourism by focusing on the impact of CPU on corporate investment for tourism and hospitality firms. Prior literature typically focuses on economic policy uncertainty (e.g., Akron et al., 2020; Demir & Gözgör, 2018), and geopolitical risk (e.g., Demir et al., 2019; Tiwari et al., 2019), with relatively less emphasis on climate policy uncertainty. We apply real options theory in a tourism context by focusing on climate policy uncertainty. Second, we extend the literature examining the impact of policy uncertainty on corporate behaviors for tourism firms by focusing on climate policy uncertainty (e.g., Akron et al., 2020; Demir et al., 2019; Demir & Gözgör, 2018; C. C. Lee & Chen, 2021). To the best of our knowledge, this is the first empirical study that examines the impact of CPU on corporate investment decisions of tourism and hospitality firms. In particular, we show the differential impact of CPU on capital and R&D investment as well as the incremental role of CPU beyond economic policy uncertainty.

We believe that the impact of CPU on corporate behaviors would be a fruitful research avenue. We only focus on the impact of CPU on corporate investment in the U.S. Considering the GDP contribution of the tourism sector at a global scale, future research could examine how climate change policy influences corporate investment at an international level, especially focusing on countries that are highly dependent on tourism. Another possible avenue for future research is to focus on the impact of climate policy uncertainty on public investment rather than private investment considered in the present study. Finally, we explore only the effects of CPU on corporate investment, one important dimension of corporate behaviors. A promising avenue for future research is to investigate how CPU affects other types of behaviors of tourism and hospitality firms.

Our study is not without limitations. First, our measure of climate policy uncertainty is obtained from Gavriilidis (2021) and constructed at the country level. However, this data may obscure large variations between states, because firms located in different regions may face varying level of climate policy uncertainty given that climate policy regulation in the U.S. is typically enacted at the state level. Therefore, future research may re-examine this important issue using a finer measure of climate policy uncertainty. Undoubtedly, our results can be generalizable to tourism and hospitality firms in other developed counties. However, it remains unclear whether our results have external validity to the tourism and hospitality sector in developing countries. Future studies may test this relation in a developing country setting.

Footnotes

Declaration of Conflicting Interests

The author(s) declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: Jing Gao acknowledges the financial support from National Natural Science Foundation of China (NSFC No.71803139).

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Lei Zhang

1.

See https://www.weforum.org/agenda/2020/01/top-global-risks-report-climate-change-cyberattacks-economic-political/ and for more details.

2.

Tourism accounted for 10.4% of the total U.S. GDP and created one in every four new jobs in 2019 (WTTC, 2021).

3.

See Section 3.2 for more detailed discussions of the CPU index.

4.

Some studies regard regulatory risk or liability risk as additional risks brought about by climate change. Specifically, transition risk is likely to be driven by regulatory risk, while liability risk refers to risk due to inadequate of disclosure of firms’ exposure to climate-related risks (In et al., 2022).

5.

We have also performed the Hausman (1978) test to determine whether random effect model or fixed effect model is more appropriate for our research setting. We find that the values of χ² are 31.78 (p-value .00) and 15.27 (p-value .02) for the capital and R&D investment regressions, respectively. Therefore, these results imply that fixed-effect estimator is superior to the random-effect estimator, consistent with prior literature (e.g., Gulen & Ion, 2016).

6.

The impact of a one standard deviation increase in CPU on capital investment is calculated as: −0.0251(coefficient reported in Table 4) × 0.58(standard deviation of CPU as reported in )/0.093(mean of capital investment as reported in Table 2) = −15.7%.

7.

We follow Tang and Jang (2009) and divide the tourism and hospitality industry into four sub-industries. Aratuo and Etienne (2019) partition the tourism sector into six sub-industries. Untabulated results suggest that our main findings continue to hold using method.

8.

It is important to note that the exclusion condition is not mathematically testable (Roberts & Whited, 2013) and caution should be exercised when interpreting this result.

9.

We also investigate the impact of COVID-19 by generating an indicator variable Covid, which takes the value of one in year 2020 and zero otherwise. Untabulated results suggest that the findings are qualitatively similar to those reported in this section.

10.

See for more details.

11.

Howe et al. (2015) show that climate change mitigation and adaptation initiatives depend on and vary with political ideology. In a similar vein, document that firms located in left-leaning states are more likely to be environmental friendly.

Author Biographies

Jing Gao is Associate Dean and Associate Professor in the International School of Economics and Management, Capital University of Economics and Business, China. Her research interests include tourism development, the relationship between tourism and energy consumption, and the relationship between climate change on tourism growth.

Lei Zhang is an Assistant Professor in Accounting at International Business School Suzhou, Xi’an Jiaotong-Liverpool University. His research interests include the impact of climate change on tourism corporate financial decision and financial reporting behaviors.

References

Abel

A. B.

(1983). Optimal investment under uncertainty. The American Economic Review, 73(1), 228–233.

Akadiri

S. S.

Alola

A. A.

Uzuner

(2020). Economic policy uncertainty and tourism: Evidence from the heterogeneous panel. Current Issues in Tourism, 23(20), 2507–2514.

Akron

Demir

Díez-Esteban

J. M.

García-Gómez

C. D.

(2020). Economic policy uncertainty and corporate investment: Evidence from the US hospitality industry. Tourism Management, 77, 104019.

Alam

M. S.

Paramati

S. R.

(2017). The dynamic role of tourism investment on tourism development and CO2 emissions. Annals of Tourism Research, 66, 213–215.

Albaladejo

I. P.

Martínez-García

M. P.

(2015). An R&D-based endogenous growth model of international tourism. Tourism Economics, 21(4), 701–719.

Arabadzhyan

Figini

García

González

M. M.

Lam-González

Y. E.

León

C. J.

(2021). Climate change, coastal tourism, and impact chains: A literature review. Current Issues in Tourism, 24(16), 2233–2268.

Aratuo

D. N.

Etienne

X. L.

(2019). Industry level analysis of tourism-economic growth in the United States. Tourism Management, 70, 333–340.

Baker

S. R.

Bloom

Davis

S. J.

(2016). Measuring economic policy uncertainty. The Quarterly Journal of Economics, 131(4), 1593–1636.

Baker

S. R.

Bloom

Davis

S. J.

Kost

K. J.

Sammon

M. C.

Viratyosin

(2020). The unprecedented stock market impact of COVID-19 (No. w26945). National Bureau of Economic Research.

10.

Banerjee

Siebert

(2017). Dynamic impact of uncertainty on R&D cooperation formation and research performance: Evidence from the bio-pharmaceutical industry. Research Policy, 46(7), 1255–1271.

11.

Barreca

Clay

Deschênes

Greenstone

Shapiro

J. S.

(2015). Convergence in adaptation to climate change: Evidence from high temperatures and mortality, 1900-2004. American Economic Review, 105(5), 247–251.

12.

Barreca

Clay

Deschenes

Greenstone

Shapiro

J. S.

(2016). Adapting to climate change: The remarkable decline in the US temperature-mortality relationship over the twentieth century. Journal of Political Economy, 124(1), 105–159.

13.

Bartram

S. M.

Hou

Kim

(2022). Real effects of climate policy: Financial constraints and spillovers. Journal of Financial Economics, 143(2), 668–696.

14.

Becken

Loehr

(2022). Tourism governance and enabling drivers for intensifying climate action. Journal of Sustainable Tourism. Advance online publication. https://doi.org/10.1080/09669582.2022.2032099

15.

Becken

Whittlesea

Loehr

Scott

(2020). Tourism and climate change: Evaluating the extent of policy integration. Journal of Sustainable Tourism, 28(10), 1603–1624.

16.

Blanford

G. J.

(2009). R&D investment strategy for climate change. Energy Economics, 31, S27–S36.

17.

Bouri

Iqbal

Klein

(2022). Climate policy uncertainty and the price dynamics of green and brown energy stocks. Finance Research Letters, 47, 102740.

18.

Brida

J. G.

Cortes-Jimenez

Pulina

(2016). Has the tourism-led growth hypothesis been validated? A literature review. Current Issues in Tourism, 19(5), 394–430.

19.

Bushee

B. J.

(1998). The influence of institutional investors on myopic R&D investment behavior. Accounting Review, 73, 305–333

20.

Chan

K. F.

Malik

(2022). Climate policy uncertainty and the cross-section of stock returns (SSRN 4075528). SSRN.

21.

Demir

Díez-Esteban

J. M.

García-Gómez

C. D.

(2019). The impact of geopolitical risks on cash holdings of hospitality companies: Evidence from emerging countries. Journal of Hospitality and Tourism Management, 39, 166–174.

22.

Demir

Gözgör

(2018). Does economic policy uncertainty affect Tourism? Annals of Tourism Research, 69(C), 15–17.

23.

Denis

D. J.

Sibilkov

(2010). Financial constraints, investment, and the value of cash holdings. The Review of Financial Studies, 23(1), 247–269.

24.

Di Giuli

Kostovetsky

. (2014). Are red or blue companies more likely to go green? Politics and corporate social responsibility. Journal of Financial Economics, 111(1), 158–180.

25.

Dillimono

H. D.

Dickinson

J. E.

(2015). Travel, tourism, climate change, and behavioral change: Travelers’ perspectives from a developing country, Nigeria. Journal of Sustainable Tourism, 23(3), 437–454.

26.

Dixit

A. K.

Pindyck

R. S.

(1994). Investment under uncertainty. Princeton University Press.

27.

Dogru

Marchio

E. A.

Bulut

Suess

(2019). Climate change: Vulnerability and resilience of tourism and the entire economy. Tourism Management, 72, 292–305.

28.

Ernst&Young. (2016). Climate change. The investment perspective. https://assets.ey.com/content/dam/ey-sites/ey-com/en_gl/topics/banking-and-capital-markets/ey-climate-change-and-investment.pdf

29.

Eugenio-Martin

J. L.

Campos-Soria

J. A.

(2010). Climate in the region of origin and destination choice in outbound tourism demand. Tourism Management, 31(6), 744–753.

30.

Farzanegan

M. R.

Gholipour

H. F.

Feizi

Nunkoo

Andargoli

A. E.

(2021). International tourism and outbreak of coronavirus (COVID-19): A cross-country analysis. Journal of Travel Research, 60(3), 687–692.

31.

Gavriilidis

(2021). Measuring climate policy uncertainty (SSRN 3847388). SSRN.

32.

Glass

L. M.

Newig

(2019). Governance for achieving the Sustainable Development Goals: How important are participation, policy coherence, reflexivity, adaptation and democratic institutions? Earth System Governance, 2, 100031.

33.

Goldstein

Koijen

R. S.

Mueller

H. M.

(2021). COVID-19 and its impact on financial markets and the real economy. The Review of Financial Studies, 34(11), 5135–5148.

34.

Gulen

Ion

(2016). Policy uncertainty and corporate investment. The Review of Financial Studies, 29(3), 523–564.

35.

Hartman

(1972). The effects of price and cost uncertainty on investment. Journal of Economic Theory, 5(2), 258–266.

36.

Hausman

(1978). Specification tests in econometrics. Econometrica, 46(6), 1251–1271.

37.

Hoang

(2022). How does corporate R&D investment respond to climate policy uncertainty? Evidence from heavy emitter firms in the United States. Corporate Social Responsibility and Environmental Management, 29, 936–949.

38.

Howe

P. D.

Mildenberger

Marlon

J. R.

Leiserowitz

(2015). Geographic variation in opinions on climate change at state and local scales in the USA. Nature Climate Change, 5(6), 596–603.

39.

S. Y.

Weyant

J. P.

Manav

(2022). Pricing climate-related risks of energy investments. Renewable and Sustainable Energy Reviews, 154, 111881.

40.

Jens

C. E.

(2017). Political uncertainty and investment: Causal evidence from US gubernatorial elections. Journal of Financial Economics, 124(3), 563–579.

41.

Julio

Yook

(2012). Political uncertainty and corporate investment cycles. The Journal of Finance, 67(1), 45–83.

42.

Kaján

Saarinen

(2013). Tourism, climate change and adaptation: A review. Current Issues in Tourism, 16(2), 167–195.

43.

Kaplan

S. N.

Zingales

(1997). Do investment-cash flow sensitivities provide useful measures of financing constraints? The Quarterly Journal of Economics, 112(1), 169–215.

44.

Kanagaretnam

Lobo

Zhang

(2022). Relationship between climate risk and physical and organizational capital. Management International Review, 29, 936–949.

45.

Khan

C. W.

Xiao

Y. D.

Zhu

Zhang

(2021). Trends in tourism under economic uncertainty. Tourism Economics, 27(4), 841–858.

46.

Kim

Wilemon

(2002). Focusing the fuzzy front–end in new product development. R&D Management, 32(4), 269–279.

47.

Lee

C. C.

Chen

M. P.

(2021). Do country risks matter for tourism development? International evidence. Journal of Travel Research, 60(7), 1445–1468.

48.

Lee

(2022). Corporate social responsibility and COVID-19: Research implications. Tourism Economics, 28(4), 863–869.

49.

Marlon

Fine

Leiserowitz

(2017). Majorities of Americans in every state support participation in the Paris Agreement (p. 8). Yale Program on Climate Change Communication.

50.

Martín

M. B. G.

(2005). Weather, climate and tourism a geographical perspective. Annals of Tourism Research, 32(3), 571–591.

51.

McCarty

N. M.

Poole

K. T.

Rosenthal

(1997). Income redistribution and the realignment of American politics. AEI Press.

52.

McCright

A. M.

Dunlap

R. E.

(2011). Cool dudes: The denial of climate change among conservative white males in the United States. Global Environmental Change, 21(4), 1163–1172.

53.

Nguyen

C. P.

Thanh

S. D.

Nguyen

(2022). Economic uncertainty and tourism consumption. Tourism Economics, 28, 920–941.

54.

OECD. (2022). Consumer confidence index (CCI) (indicator). Author. https://doi.org/10.1787/46434d78-en

55.

Orfila-Sintes

Crespi-Cladera

Martínez-Ros

(2005). Innovation activity in the hotel industry: Evidence from Balearic Islands. Tourism Management, 26(6), 851–865.

56.

Pablo-Romero

M. D. P.

Molina

J. A.

(2013). Tourism and economic growth: A review of empirical literature. Tourism Management Perspectives, 8, 28–41.

57.

Paramati

S. R.

Alam

M. S.

Lau

C. K. M.

(2018). The effect of tourism investment on tourism development and CO2 emissions: Empirical evidence from the EU nations. Journal of Sustainable Tourism, 26(9), 1587–1607.

58.

Petersen

M. A.

(2009). Estimating standard errors in finance panel data sets: Comparing approaches. The Review of Financial Studies, 22(1), 435–480.

59.

Roberts

M. R.

Whited

T. M.

(2013). Endogeneity in empirical corporate finance1. In Constantinides

G. M.

Harris

Stulz

R. M.

(Eds.) Handbook of the economics of finance (Vol. 2, pp.493–572). Elsevier.

60.

Romer

P. M.

(1990). Endogenous technological change. Journal of Political Economy, 98(5, Part 2), S71–S102.

61.

Sautner

van Lent

Vilkov

Zhang

(2020). Firm-level climate change exposure (European Corporate Governance Institute–Finance Working paper 686). European Corporate Governance Institute–Finance.

62.

Schliephack

Dickinson

J. E.

(2017). Tourists’ representations of coastal managed realignment as a climate change adaptation strategy. Tourism Management, 59, 182–192.

63.

Scott

(2011). Why sustainable tourism must address climate change. Journal of Sustainable Tourism, 19(1), 17–34.

64.

Scott

Gössling

Hall

C. M.

(2012). International tourism and climate change. Wiley Interdisciplinary Reviews: Climate Change, 3(3), 213–232.

65.

Segal

Shaliastovich

Yaron

(2015). Good and bad uncertainty: Macroeconomic and financial market implications. Journal of Financial Economics, 117(2), 369–397.

66.

Selvanathan

E. A.

Jayasinghe

Selvanathan

(2022). International tourism and infectious disease transmission nexus: A cross-country and regional study. Journal of Travel Research, 61(8), 1910–1927.

67.

Standard and Poor’s. (2017). How environmental and climate risks and opportunities factor into global corporate ratings: An update. https://www.spratings.com/documents/20184/1634005/How+Environmental+And+Climate+Risks+And+Opportunities+Factor+Into+Global+Corporate+Ratings+-+An+Update/5119c3fa-7901-4da2-bc90-9ad6e1836801

68.

Steiger

Scott

Abegg

Pons

Aall

(2019). A critical review of climate change risk for ski tourism. Current Issues in Tourism, 22(11), 1343–1379.

69.

Stock

J. H.

Yogo

(2005). Testing for weak instruments in linear IV regression. In Donald

W. K. A.

(Ed.), Identification and inference for econometric models (pp. 80–108). Cambridge University Press.

70.

Tang

C. H. H.

Jang

S. S.

(2009). The tourism–economy causality in the United States: A sub-industry level examination. Tourism Management, 30(4), 553–558.

71.

Tiwari

A. K.

Das

Dutta

(2019). Geopolitical risk, economic policy uncertainty and tourist arrivals: Evidence from a developing country. Tourism Management, 75, 323–327.

72.

Hartley

(2022). Drivers of growth and catch-up in the tourism sector of industrialized economies. Journal of Travel Research, 61(5), 1156–1172.

73.

Wilson

N. E.

(2021). The impact of competition on investment: Evidence from California hospitals. The Journal of Industrial Economics, 69(1), 1–32.

74.

World Economic Forum. (2020). Global risks report 2020: These are the top risks facing the world. Retrieved January 12, 2022, from https://www.weforum.org/agenda/2020/01/top-global-risks-report-climate-change-cyberattacks-economic-political/

75.

World Economic Forum. (2021). These are the world’s greatest threats in 2021. Retrieved January 12, 2022, from https://www.weforum.org/agenda/2021/01/these-are-the-worlds-greatest-threats-2021/

76.

World Travel and Tourism Council. (2021). United States: 2021 annual research: Key highlights. Author.

77.

Zhang

Gao

(2016). Exploring the effects of international tourism on China’s economic growth, energy consumption and environmental pollution: Evidence from a regional panel analysis. Renewable and Sustainable Energy Reviews, 53, 225–234.