International Journal of Hospitality Management 78 (2019) 169–178 Contents lists available at ScienceDirect International Journal of Hospitality Management journal homepage: www.elsevier.com/locate/ijhm Asset-light business model: An examination of investment-cash flow sensitivities and return on invested capital T Kwanglim Seoa, , Jungtae Sohb ⁎ a b School of Travel Industry Management, University of Hawaii, Manoa, United States School of Hospitality Management, The Pennsylvania State University, United States A R TICL E INFO A BSTR A CT Keywords: Asset-light business model Investment Cash flow Return on invested capital Lodging industry The purpose of this study is to investigate the effects of the asset-light business model on investment-cash flow sensitivities and return on invested capital in the lodging industry. Little research has explored the link between investment and asset ownership structure. The current study provides an alternative approach to examining investment behavior and return on invested capital, focusing on the unique characteristics of asset ownership among lodging firms. The findings of this study provide important implications for lodging investors and shareholders regarding the strategic use of the asset-light business model for aiding lodging firms’ efficient investments and delivering high return on invested capital. 1. Introduction Recently, major lodging firms have increasingly reduced their ownership of hotel assets and moved to an asset-light business model. An asset-light business model (ALBM) refers to a business model that concentrates more on managing and franchising hotels rather than owning and controlling the real estate, enabling lodging firms to develop and operate hotels with little or no capital investment (Sohn et al., 2013). Under this business model, lodging firms can achieve scale and earnings growth by adding new hotels to the system without having to incur large initial capital expenditures. Minimizing the need for large capital investment allows lodging firms to become more flexible in their investment decisions, suggesting that the dynamics of investment may vary among lodging firms according to the level of the asset ownership. Fazzari et al. (1988) argued that the investment of a firm is generally sensitive to the availability of cash flow especially when the firm has restricted access to external capital markets (e.g. financially constrained). Financially constrained situations arise from the unavailability of external funds for the firm’s investment, requiring the firm’s dependency on the availability of internal funds. As a result, a financially constrained firm becomes very sensitive to changes in cash flows for investing decisions. For instance, the firm is more likely to reduce its investment spending when investment needs to be mainly financed internally than when external funds are available. However, recent studies identified numerous problems, which question the mechanism that drives the relationship between investment and cash flow (Almeida ⁎ and Campello, 2007; Alti, 2003; Erickson and Whited, 2000; Kaplan and Zingales, 1997, 2000; Li and Tang, 2008). To provide an alternative explanation for this relationship, the current study explores the idea that a specific business model in the lodging industry will have a significant impact on corporate investments. In particular, because ALBM greatly reduces requirements for capital investment in comparison to the traditional business model, this study argues that ALBM could alleviate the investment-cash flow sensitivity (ICFS). Since financially constrained firms are more sensitive to available cash flow, this moderating effect is expected to be more prominent when firms face financial constraints. As the core of ALBM is to develop a firm’s competitive advantages by reducing its own investment and efficiently allocating its capital to the most profitable investments (Maly and Palter, 2002), another fundamental question to answer is whether or not ALBM can improve the efficiency of lodging firm’s capital investment. With significant amounts of unencumbered capital, otherwise assigned to expensive real estate, lodging firms can effectively invest in the development of core competencies to improve the efficiency and quality of hotel operations, leading to greater returns on investments. Therefore, this study argues that ALBM will aid lodging firms in achieving competitive advantages in return on invested capital (ROIC). Given that core competencies and competitive advantages vary among different hotel segments, this study further examines the effect of ALBM on segments representing fullservice and limited-service hotels. While understanding the extent to which asset structures affect Corresponding author. E-mail addresses: kwanglim@hawaii.edu (K. Seo), jks5501@psu.edu (J. Soh). https://doi.org/10.1016/j.ijhm.2018.12.003 Received 22 February 2018; Received in revised form 3 December 2018; Accepted 4 December 2018 0278-4319/ © 2018 Elsevier Ltd. All rights reserved. International Journal of Hospitality Management 78 (2019) 169–178 K. Seo, J. Soh about investment opportunities, generating the cross-sectional patterns across firms in ICFS regardless of financing constraints. Similarly, Erickson and Whited (2000) showed that variations in ICFS among financially constrained and unconstrained firms may result from measurement error in investment opportunities. Moyen (2004) also argues a potential correlated omitted variable problem that can cause a positive relationship between investment and cash flow without financing frictions. Despite extensive research efforts, the findings of the extant studies may not apply directly to the lodging industry as their sample was limited to manufacturing firms, suggesting industry-specific research is needed to test ICFS in the hospitality industry. While there are a limited number of studies addressing ICFS in the hospitality literature, empirical evidence is still debatable. Examining hotel real estate investment trusts (REITs) and hotel C-corporations, Kim and Jang (2012) found that ICFS is greater for hotel REITs because of a lack of internal funds after mandatory dividend payments to shareholders. However, contrary to expectations that investments of financially constrained small restaurant firms would be more sensitive to cash flow, Upneja and Sharma (2009) showed that relatively unconstrained large restaurant firms displayed higher degrees of ICFS. While the findings of these studies are inconsistent, the effect of financing constraint was not explicitly considered in the hospitality literature as they did not empirically measure the level of financing constraint for firms. Therefore, using three a priori measures of financial constraint, this study constructs subsamples of financially constrained and unconstrained lodging firms, and compares ICFS across the subsamples. Consistent with the proposition by Fazarri et al., the current study proposes the following hypothesis. investment is important, the extant literature has not considered the relationship between them. Focusing on the unique characteristic of asset ownership among lodging firms, the present study provides an alternative approach to examining investment behavior and ROIC. When investment demand is influenced by a firm’s specific asset strategy, firms adopting ALBM will respond to cash flow shocks differently and choose an investment policy that is contrary to that of firms that should self-fund all of their investments. The difference in the investment policies of these firms further allows us to formulate an empirical prediction about the effects of asset-light strategies on the efficiency of investment to generate a superior return. To our knowledge, our paper is the first to explore the link among investment, ROIC, and assets’ ownership structures. The primary objectives of this study are to (1) investigate the effect of ALBM on the relationship between investment and cash flow, and (2) examine the effect of ALBM on ROIC in the lodging industry. The findings of this study provide important implications for lodging investors and shareholders that the strategic use of ALBM can help lodging firms’ efficient investments and deliver high ROIC. 2. Literature review 2.1. Financing constraints and ICFS While investment decisions revolve around how to best allocate capital to maximize return, financing decisions help find the optimal way to finance the investment. The finance literature documents how financing decisions are made under imperfect market conditions where the cost of capital to the firm varies with respect to financing sources (Kaplan and Zingales, 1997). The pecking order theory states that a firm prefers to finance investments with internal funds over external funds because of higher external financing costs, suggesting that the firm’s investment may become sensitive to the availability of internal cash flows when its ability to access external financing is constrained (Myers and Majluf, 1984). In particular, theories of financing constraints predict that the firm’s investment decisions are largely dependent on its accessibility to external capital markets while it is subject to financing constraints (Calomiris and Hubbard, 1995). Financing constraints are referred to as restrictions, such as lack of collateral, excessive leverage, credit constraints, and information asymmetries, that increase the expected costs of external financing (Kaplan and Zingales, 1997). For instance, firms with little or no collateral are likely to face more substantial external financing costs by lenders. Hence, Fazzari et al. (1988) argued, in the presence of financial constraints, firms rely more heavily on internal funds for investment, and their investment decisions become highly sensitive to the availability of internal cash flows due to the significant cost difference between internal and external funds. In turn, financially unconstrained firms’ investment decisions are independent of cash flow because they have unlimited access to external capital markets. Following an influential study of Fazzari et al. (1988), many scholars tested ICFS on two groups of firms classified as financially unconstrained and constrained based on various financial characteristics those firms with high (low) dividend payout ratio, large (small) asset size, and their bonds rated (unrated) are considered financially unconstrained (constrained). Empirical evidence shows that investment decisions of financially constrained firms are highly sensitive to changes in cash flow when compared to financially unconstrained firms (Abel and Eberly, 2011; Brown and Petersen, 2009; Calomiris and Hubbard, 1995; Fazzari and Petersen, 1993; Fazzari et al., 2000; Himmelberg and Petersen, 1994; Lewellen and Lewellen, 2016). However, other scholars questioned a theoretical and empirical model by Fazzari et al., arguing that ICFS is not a useful indicator of financial constraints (Chen and Chen, 2012; Gala and Gomes, 2013; Hadlock and Pierce, 2010; Kaplan and Zingales, 1997). For example, Alti (2003) and Gomes (2001) argued that cash flows can also contain information H1. ICFS is higher for financially constrained lodging firms than unconstrained lodging firms. 2.2. The effect of ALBM on ICFS As discussed above, the predictions of financing constraints still remain controversial in the literature, encouraging scholars to explore alternative explanations for the relationship between investment and cash flow. Several scholars attempted to explain this relationship by including the interaction terms that capture the effect of strategic decisions on ICFS (Brown and Petersen, 2009; Khramov, 2012). In this paper, we argue that ALBM that mitigates investment needs and capital requirements may also moderate the relationship between investment and cash flow – that is, investment decisions of asset-light lodging firms are less susceptible to changes in cash flow as they are not responsible for the costs and all capital expenditures (Kim et al., 2019). Hence, the asset-light strategy allows greater flexibility to respond to investment opportunities. From the real estate management perspective, lodging firms holding real estate assets are exposed to risks associated with real estate investments, such as depreciation, high leverage, high opportunity costs, and low liquidity (Deng et al., 2017; Hoesli and Oikarinen, 2012; Kim et al., 2019; Tuzel, 2010). Page (2007) maintained that the asset-light strategy can effectively reduce real estate risk, allowing firms to be more flexible when making investment decisions and allocations. Several lodging firms also describe the impact of ALBM in their annual reports. Hyatt Hotels Corp. (2017) states that increased risks and costs resulting from significant capital investments in owned hotels could limit their ability to capitalize on investment opportunities. Marriott International Inc. (2016) also notes that minimizing capital investments allows them to maximize financial flexibility and control because they don’t absorb the full impact of fluctuating profits on cash flows. These arguments suggest that the asset-light firms don’t have the same level of sensitivity of investment to cash flow compared to lodging firms with greater property ownership. With regards to asset ownership, Almeida and Campello (2007) 170 International Journal of Hospitality Management 78 (2019) 169–178 K. Seo, J. Soh argued that asset tangibility could influence ICFS when the firm has constrained access to external financial resources. In general, the level of investment is dependent on the availability of both credit and cash flow (Almeida et al., 2004). Since lenders’ asset valuation for collateral determines the constraint on a firm’s credit, high asset tangibility allows the firm to increase borrowing capacity for investment based on the high collateral value of assets (Hovakimian and Titman, 2006). Using a large sample of U.S. manufacturing firms from 1985 to 2000, Almeida and Campello (2007) found that high asset tangibility through increased borrowing capacity intensifies the positive effect of cash flow on investment spending of financially constrained firms. That is, an increase in investment following a positive cash flow shock will be greater for the firm that invests in highly tangible assets due to increased debt capacity. On the other hand, they maintained that the firm becomes financially unconstrained when asset tangibility is very high, and consequently, ICFS becomes independent of asset tangibility. However, due to reduced capital requirements and greater financial flexibility, ICFS of lodging firms pursuing ALBM may be less relevant to asset tangibility. ALBM helps improve profitability and produce stable earnings, allowing lodging firms to establish high credit ratings and increase capacity for debt (Roh, 2002; Sohn et al., 2013). For instance, major lodging firms pursuing the asset-light strategy maintained investment-grade credit ratings based on the positive outlook for high earnings growth.1 In addition, lodging firms pursuing asset-light structures show lower volatility in profit as their costs are more variable to their revenues. Considering different business cycles, Sohn et al. (2014) found that earnings of fee-based hotel firms are less volatile during periods of contracting markets while these hotels are more likely to grow during periods of expansion. Therefore, unlike previous findings, the current study predicts that for asset-light lodging firms, ICFS will not increase according to levels of asset tangibility. Instead, ALBM will negatively moderate the relationship between investment and cash flow, alleviating ICFS. Furthermore, because the investment becomes more sensitive to changes in internal funds when firms encounter constraints to access capital markets, the negative moderating effect of ALBM becomes stronger for financially constrained firms than unconstrained firms. In other words, for firms highly positioned as asset-light, investment will not be very sensitive to the available cash flow even when financially constrained. Therefore, we developed the following hypotheses: Wiersema, 1999, p.628–629). That is, efficient allocation and utilization of scarce resources is the core aspect of developing competitive advantages that ultimately enhance a firm’s value. From the resourcebased perspective, ALBM can help establish competitive advantages both directly and indirectly. First, asset-light lodging firms can gain cost efficiency advantages directly by outsourcing capital investment responsibilities (Hansen et al., 2009; Simon, 2015). Minimizing the commitment to fixed assets can lead to significant cost savings from reductions in the cost of development. Empirical evidence shows that firms engaging in outsourcing activities accomplish greater cost flexibility and efficiency than non-outsourcing firms (Gonzalez et al., 2013; Jiang et al., 2006). In addition, ALBM can deliver higher cost efficiency gains by achieving scalability. For instance, an asset-light firm can increase the number of hotels in its network without investing capital and reach to achieve economies of scale. Capitalizing on the scale helps firms increase revenue with minimal incremental cost while significantly reducing operating costs through centralized and streamlined processes (EspinoRodríguez and Lai, 2014). Second, an indirect effect of low capital requirements creates available capital to focus on strengthening core competencies (Rothaermel and Hess, 2007). That is, a reduction in the level of fixed asset investment assists firms in accumulating capital for advantageous allocation to the development of core competencies that create superior value to customers (Dennis, 2011; Prahalad and Hamel, 1990). For instance, saved capital, can be efficiently invested in the core business of managing and franchising hotels to build effective marketing, property management, revenue management, and reservation systems, rigorous training and educational programs, and enhancement through updated information technologies, all of which contribute to establishing strong brands, customer relations, human resources, and delivery of excellent services and products, and eventually superior performance. By adopting ALBM, Hyatt generated savings to fund growth and build their brands while providing their hotels with advanced tools and analytics from regional and corporate offices (Hyatt Hotels Corp, 2017). In the hospitality literature, research efforts have been mainly undertaken to investigate the relationship between the asset-light strategy and performance. Although there is an implicit assumption that ALBM assures optimal performance for all lodging firms, empirical evidence has been somewhat inconclusive. Using path analysis, Sohn et al. (2013) tested a mediation effect of operating profitability, finding that an asset-light and fee-oriented strategy positively influences operating profitability and a firm’s value in the U.S. hotel industry. Other studies, however, revealed that the implementation of ALBM had no significant or limited effect on the performance of lodging firms (Blal and Bianchi, 2019; Yu and Liow, 2009). In particular, by investigating the role of hotel properties in mixed asset portfolios, Low et al. (2015) found that portfolios with asset-intensive hotel firms outperformed those with asset-light hotel firms, suggesting financial performance should be carefully evaluated before implementing ALBM. The mixed findings call for a deeper examination of more direct consequences of the implementation of ALBM – reallocation of resources saved from divestment of real estate properties. Therefore, using ROIC, this study explicitly investigates an asset-light firm’s efficiency at allocating the resources to profitable investments. Strategic management literature argues that ROIC can show how efficiently a firm utilizes resources to generate returns (Damodaran, 2007; Tang and Liou, 2010). In particular, the current study predicts that the efficiency of invested capital would increase as the level of asset-lightness increases. This study proposes the following hypothesis. H2(a). ALBM will negatively moderate the relationship between investment and cash flow. H2(b). The negative moderating effect of ALBM is more prominent for financially constrained firms than unconstrained firms. 2.3. The effect of ALBM on ROIC Assuming its influence on investment decisions, whether or not ALBM leads to superior returns is another important empirical question. According to the resource-based view, “a firm’s performance depends fundamentally on its ability to have a distinctive, sustainable competitive advantage which derives from the possession of unique, nonimitable, non-transferable, firm-specific resources” (Bowen and 1 The two largest credit rating agencies, Standard & Poor’s and Moody’s, publish the credit ratings that represent the quality of corporate or government bonds (Kim and Gu, 2004). These agencies use letter designations such as A, BB, CCC to describe the quality of a bond. In particular, bond credit ratings of BBBor higher by Standard & Poor’s or Baa3 or higher by Moody’s are considered investment grade, indicating that the issuer of the bond is likely to meet its financial obligations. Most of the sampled lodging firms maintained investment grade credit ratings during the study period. For instance, Marriott International Inc. has maintained an acceptable rating for investment with Standard & Poor’s and Moody’s since 1998. H3. ALBM has a positive impact on ROIC for lodging firms. 171 International Journal of Hospitality Management 78 (2019) 169–178 K. Seo, J. Soh 2.4. The effect of ALBM across full-service and limited-service segments 3. Methodology In the lodging industry, hotel asset classes generally represent two main segments, full- and limited-service segments, based on the nature of services and the target market (Enz, 2010). Full-service hotels typically face higher market expectations and greater capital requirements than limited-service hotels (Enz et al., 2013; Kim et al., 2013). For instance, compared to limited-service hotels, full-service hotels are not only expensive to develop but require frequent periodic capital investments for upgrades and improvements to sustain the high standards of quality and services. Furthermore, given the complicated operating structure of a full-service hotel, its fixed cost structure is higher than that of a limited-service hotel (Younes and Kett, 2007). Therefore, ALBM may present greater competitive advantages for full-service hotels because money saved on capital expenditures and fixed costs is more considerable than limited-service hotels. In addition, factors affecting purchase decisions of consumers vary considerably between full- and limited-service segments, suggesting that lodging firms should develop distinct business and investment strategies for each segment (Chu and Choi, 2000; Kim et al., 2013; Peng et al., 2015; Tanford et al., 2011). Typically, full-service hotels’ customers are less price-conscious and have higher service quality expectations than do limited-service hotels’ guests (Lai et al., 2014; Lee, 2014; Peng et al., 2015; Zhang et al., 2013). A recent study showed that nonprice related attributes, such as emotional connections, more likely influence full-service hotels’ guests’ purchases while price and valuerelated attributes influence patrons of limited-service hotels (Tanford et al., 2012). Other research further suggests that full-service hotels can take advantage of expertise, experience, and knowledge that sophisticated management provides because products and services in the fullservice segment involve higher levels of operational complexity than the limited-service segment (Lee, 2015; Silva, 2015; Wu et al., 2012). The findings of these studies imply that lodging firms should focus their resources on investing in the development of their core competencies to meet specific customer needs and expectations in a target segment. For instance, hotels operating primarily in the full-service segment can benefit from developing specialized marketing and loyalty programs to strengthen consumer confidence and commitment to their brands. Enz et al. (2013) examined new hotels entering the market in the U.K., finding that brand affiliated full-service hotels obtained higher levels of RevPAR in the first six months of operations while brand affiliated limited-service hotels earned no initial performance advantages.2 Therefore, Competitive advantages achieved by ALBM will create greater value for lodging firms developing and operating more properties in the full-service segment than the limited-service segment. Although the extant studies provide some useful information regarding the benefits of ALBM, prior research does not consider its direct and indirect impacts on the efficiency of investment across different market segments in the U.S. lodging industry. In particular, the current study predicts that due to superior cost efficiency and the development of core competencies, lodging firms will benefit from implementing ALBM when developing and operating more full-service hotels while the benefit may not be significant when focusing on limited-service hotels. This study proposes the following hypotheses. 3.1. Sample and data collection The sample of this study consists of publicly traded U.S. lodging firms using data during 1998–2018. The study’s period is due to most U.S. public lodging firms having begun to provide detailed property information in financial reports since 1998. Collection of quarterly financial data is from the Compustat database based on the Standard Industrial Classification (SIC) code of Hotels and Motels (7011). Quarterly reports (10Q) and annual reports (10 K) provided detailed managerial and franchising information, such as the number of managed and franchised hotels, the number of full-service and limitedservice hotels, and the total number of hotels. Winsorizing all variables at the 1st and 99th percentiles on a quarterly basis reduced the effect of outliers (Baker et al., 2003). The final sample consisted of 1,478 firmquarter observations with no missing values. 3.2. Variables 3.2.1. Dependent and independent variables This study uses two dependent variables: investment and return on invested capital (ROIC). Following Fazzari et al. (1988), investment (I) is the ratio of capital expenditures to the beginning of period capital stock. ROIC is a calculation of after-tax operating income divided by the beginning of period book value of invested capital (Damodaran, 2007). This study also includes four independent variables. Cash flow (CF) measures the availability of cash flow within a firm. Previous research of ICFS generally argued that a firm’s investment is sensitive to cash flow (Fazzari et al., 1988). CF is a measurement of income before extraordinary items plus depreciation and amortization divided by the beginning of period capital stock (Chen and Chen, 2012). The degree of asset-lightness was measured in two alternative ways. First, we measured asset lightness using the ratio of the sum of franchised and managed properties to the total number of properties (AL1). To construct this measure, we hand-collected data for franchised and managed properties using firms’ quarterly and annual reports. Second, following Liou (2011), we constructed another firm-level measure of asset lightness (AL2) by taking the ratio of light assets to tangible assets, where the value of light assets is the excess return over risk free rate multiplied by book value of assets plus goodwill and intangibles. Finally, two developed variables group lodging firms’ operations into full-service and limited-service segments. Full (FS) and limited-service (LS) variables are measurements of the ratio of the total number of full (limited)service hotels to the total number of hotels. The prediction is that the effect of AL could vary between full-service and limited-service segments. 3.2.2. Control variables Several control variables are employed in this study. First, ICFS considers Tobin’s q (Q) to control for firms’ investment opportunities (Fazzari et al., 1988; Kaplan and Zingales, 1997). Almeida and Campello (2007) argued that future demand and opportunities for investments influence a firm’s decisions. Q is the measurement of the market value of assets divided by the book value of assets. Second, larger firms generally benefit from economies of scale when compared to smaller firms (Almeida et al., 2004); therefore, firm size (SIZE), measured by the log of sales, controls for any systematic effect among firms of different sizes. Third, finance literature recognized leverage (LEV) as an important factor in investment decisions and firm performance. Johnson (2003) maintained that firms are less likely to invest when carrying high levels of debt. High leverage increases debt-related expenses, suppressing firms’ investment spending (Diamond and He, 2014). A negative relationship between leverage and firm value is likely since leverage indicates higher levels of default risk (Barclay and Smith, 1995; Guedes and Opler, 1996). Measurement of LEV is according to H4(a). The positive effect of ALBM on ROIC increases for lodging firms when operating more in the full-service segment. H4(b). The positive effect of ALBM on ROIC is not significant for lodging firms when operating more in the limited-service segment. 2 Revenue per available room (RevPAR) has general acceptance as one of the most important performance measures in the lodging industry (Enz et al., 2013). RevPAR is a calculation from dividing the revenue generated from rooms sold by the total number of rooms available for sale. 172 International Journal of Hospitality Management 78 (2019) 169–178 K. Seo, J. Soh total long-term debt divided by market value of the firm (Stohs and Mauer, 1996). items plus depreciation and amortization expense, CFit, scaled by capital stock at the beginning of the period, Kit-1; ALit indicates the degree of asset-lightness; (CF × AL)it indicates the interaction of cash flow with the degree of asset-lightness. Hypothesis 3 and 4 test for the effect of ALBM on ROIC and its effect across full- and limited-service segments. Regression Model (6) and (7) test Hypothesis 3 and 4: 3.2.3. Model specification This study developed four regression models to test proposed hypotheses. Hypotheses 1 and 2 test for ICFS relative to financial constraints and ALBM, respectively. Testing hypotheses 1 and 2 requires classifying sampled lodging firms based on a priori measures of the financial constrains they face. While there are several approaches to sorting firms in the literature, it is still debatable which approach works best (Fazzari et al., 2000; Kaplan and Zingales, 2000) In particular, researchers argue that the sample splits may vary with respect to specific set of financial features used to characterize the status of financially constrained and unconstrained. To address this concern, we used three alternative measures to separate our sample. First, using the index of financial constraints developed by Kaplan and Zingales (1997) (KZ Index, hereafter), divides firms into ten equal percentage classes (deciles). Firms ranking in the top three deciles of the KZ Index classify as financially constrained while those in the bottom three deciles classify as financially unconstrained. Construction of the KZ Index uses equation: ROICit = ROICit = 0 ALit + 1 FSit + 2 LSit + 3 SIZEit 0 ALit + 1 FSit + 2 LSit + 3 (AL + 5 SIZEit + 6 LEVit + it + 4 LEVit × FS )it + + it 4 (AL (6) × LS )it (7) where CF is cash flow, measured as the ratio of cash flow to total assets; DP is dividend policy, adopting a value of one if the firm pays dividends; LTA is the ratio of the long-term debt to total assets; TA is the natural log of total assets; SG is the firm’s sales growth, measured as the difference between sales in time t and t + 1, divided by the sales in time t. The replacement cost of total assets scale all variables. The higher (lower) the WW Index is, the more (less) financially constrained a firm is. Regression Models (4) and (5) test Hypotheses 1 and 2: where ROICit Quarterly Journal of Economics < represents return on invested capital, proxied as net income divided by the sum of interestbearing liabilities and owners’ equity; ALit is the degree of asset-lightness; FSit (LSit) represents the degree of full (limited)-service hotels in a firm’s portfolio, proxied by the ratio of the total number of full (limited)-service properties to the total number of properties; SIZEit is firm size, measured as the log of sales; LEVit is leverage, proxied as long-term debt divided by the market value of the firm. To avoid estimation biases, we employed several techniques to estimate regression Eqs. (4)–(7). First, we included firm- and time-fixed effects in all of our estimations to control for biases resulting from unobserved individual heterogeneity and time idiosyncrasies (Campa and Kedia, 2002). Using the ordinary least squares technique would produce biased coefficients due to existence of time-invariant unobservable firm- and time-specific effects correlated with other regressors in the model. For example, if a firm’s decision to go asset-light depends on the presence of some unobserved firm-specific characteristics, including the fixed-effects parameters will partial out these unobserved effects within groups. Second, we analyzed our regression models by clustering standard errors by firm. While including firm-fixed effects in all of our estimations effectively removes unobserved timeinvariant heterogeneity from error term, the remainder of the error might still be correlated over time. Clustered standard errors can account for remaining serial correlation in the dependent variable that may not be corrected by the fixed-effects model (Thompson, 2011; Wooldridge, 2002). Thirdly, to further control for potential endogeneity problems, we used a General Method of Moments (GMM) estimator proposed by Arellano and Bond., (1991) While the fixed-effects model generally produces consistent estimates of the endogenous variables, unobserved effects that are not constant over time can introduce endogeneity biases in the analysis. Following Almeida and Campello (2007), we instrumented differenced regressors by their lagged levels (i.e., the lags of the investment and the first differences of the independent variables were used as instruments for the first-differenced regression equations). Hansen’s J test was conducted to check the validity of the instruments in all regressions (Hansen, 1982). Furthermore, Regression Models (5) and (7) include interaction terms to test for the moderating effect of the asset-light strategy, suggesting potential multicollinearity problems in the regression equation. To overcome problems of multicollinearity, this study mean-centered relative variables before creating interaction terms (Aiken and West, 1991). Analysis of the variation inflation factors (VIF) from each regression model confirms the lack of significant multicollinearity issues (Tabachnick and Fidell, 1996). Iit Kit 1 4. Results and discussion Iit Kit 1 KZIndex = 1.002 × CF + 0.283 × Q + 3.139 × LEV 39.368 × DIV (1) 1.315 × CH where CF is cash flow, which is earnings before extraordinary items plus depreciation and amortization expense; Q is Tobin’s q, measured as the market value of assets divided by the book value of assets; LEV indicates firm’s leverage, proxied as long-term debt divided by the market value of the firm; DIV represents dividends, measured as the total dividend divided by the book value of total assets; CH is cash holdings, calculated by the ratio of cash and marketable securities to total assets. Second, following the study of Hadlock and Pierce (2010), the SizeAge Index (SA Index, hereafter) categorizes financially constrained and unconstrained firms. The SA Index’s calculation uses firm’s size and age in the equation: 0.737 × SIZE + 0.043 × SIZE 2 SAIndex = (2) 0.040 × AGE where measuring SIZE uses the natural log of the book value of total assets; AGE is the number of years a firm has a listed stock price on Compustat. Higher (lower) SA Index indicates that firms are financially more (less) constrained. Third, this study adopted the Whited and Wu Index (WW Index, hereafter) (Whited and Wu, 2006). The WW Index uses the equation: WWIndex = 1.091 × CF 0.062 × DP + 0.021 × LTA 0.035 × SG = = 0 + 0 + 1 qit 1 + 1 qit 1 + 2 CFit + Kit 1 it 2 CFit + Kit 1 3 ALit 0.044 × TA (3) (4) + 4 (CF × AL)it + it 4.1. Descriptive statistics (5) The summary of variables’ descriptive information appears in Table 1. The mean Q value is 1.974, suggesting existence of greaterthan-average investment opportunities for all sampled firms. To execute and capitalize on the investment opportunities, these firms should where Iit / Kit-1 is investment, measured as the ratio of capital investment, Iit, scaled by capital stock at the beginning of the period; Kit-1; qit1 is a proxy for investment opportunities, measured by Tobin’s q; CFit / Kit-1 represents cash flow, measured as earnings before extraordinary 173 International Journal of Hospitality Management 78 (2019) 169–178 K. Seo, J. Soh Table 1 Summary of descriptive statistics. Ii,t Qi,t CFi,t AL1i,t AL2i,t ROICi,t SIZEi,t LEVi,t FSi,t LSi,t Table 3 Investment-cash flow sensitivity: Financially constrained vs. unconstrained. N Mean Median S.D. Min Max 1,478 1,478 1,478 1,478 1,478 1,478 1,478 1,478 1,478 1,478 0.086 1.973 0.112 0.816 0.076 0.064 4.464 0.376 0.585 0.463 0.057 1.587 0.045 0.599 0.047 0.032 3.691 0.297 0.511 0.454 0.102 1.935 0.191 0.439 0.230 0.078 0.317 0.289 0.386 0.371 0 0.303 −1.378 0 −9.104 −0.091 0.613 0 0 0 0.842 26.105 2.461 1 5.241 0.783 7.932 2.216 1 1 Dependent variable: I (1) KZ Index Constrained Unconstrained (2) SA Index Constrained Unconstrained Note: Ii,t represents capital investment, defined as the ratio of capital expenditures to capital stock’s value at the beginning of the period. Qi,t is Tobin’s q, proxied by the market value of assets over the book value of assets. CFi,t is cash flow, defined as the sum of depreciation, amortization, and income before extraordinary items divided by capital at the beginning of the period. AL1i,t represents the degree of the asset-lightness, measured as the ratio of the sum of franchised and managed properties to the total number of properties. AL2i,t is the degree of the asset-lightness, measured as the ratio of light assets to tangible assets, where the value of light assets is the excess return over risk free rate multiplied by book value of assets plus goodwill and intangibles. ROICi,t is return on invested capital, calculated as after-tax operating income divided by the beginning of period book value of invested capital. SIZEi,t is firm size, calculated by the log of sales. LEVi,t is leverage, long-term debt divided by total assets. FSi,t represents the full-service segment’s proportion of a firm’s operations, measured as the ratio of the total number of full-service properties to the total number of properties. LSi,t represents the limited-service segment’s proportion of a firm’s operations, measured as the ratio of the total number of limited-service properties to the total number of properties. Subscripts, i and t, represent firm and quarter, respectively. (3) WW Index Constrained Unconstrained Independent variables Qi,t CFi,t N R2 J 0.005 (1.68) 0.003 (1.41) 0.246** (3.68) 0.118 (1.83) 444 .33 .58 443 .29 .30 0.008 (1.86) 0.005 (1.30) 0.388** (5.41) 0.091 (1.41) 444 .35 .77 443 .42 .48 0.006 (1.14) 0.003 (1.12) 0.286** (3.96) 0.121* (2.53) 444 .42 .41 443 .31 .29 * Significant at 0.05. ** Significant at 0.01. effects with standard errors clustered by firm. Hanesen’s J-statistics suggest that our instruments are valid. The coefficients on CF are not only positive and statistically significant in all of the constrained firm estimations (t = 3.68, 5.41, and 3.96) but also consistently greater than those of the unconstrained firms. Therefore, Hypothesis 1 receives support. Panel A and Panel B of Table 4 report the results from the estimations of Regression Model (5). As in our previous estimations, the model was estimated via GMM with firm- and year-fixed effects with standard errors clustered by firm. The degree of asset-lightness was proxied by AL1 and AL2 in columns (I) and (II), respectively. Hanesen’s J-statistics confirmed the validity of the instrumental variables used in GMM for the equation. In addition, VIF verified the absence of significant multicollinearity regarding the interaction terms used in all regressions. Hypothesis 2(a) predicts that ALBM will negatively moderate the relationship between investment and cash flow. Consistent with the argument that the investment of lodging firms pursuing the asset-light strategy is less sensitive to changes in cash flows, the coefficient on the interaction term of CF and AL1 (AL2) is negative and significant in row (2) of Panel A (t=-3.76 and -3.64). Our findings are also economically significant. For example, the estimates in row (2) of Panel A suggest that when calculated at the third quartile of AL1 (i.e., more asset-light), each dollar of cash flow decrease would be associated with an increase of 55 cents in investment. When, however, calculated at the first quartile of AL1 (i.e., less asset-light), an additional dollar drop in cash flow would be associated with an increase of only 14 cents in investment. Hence, Hypothesis 2(a) gains support, attributing an important role to the specific choices for structures for owning assets, thereby forming the investing behavior of lodging firms. consider the availability of internal cash flows and accessibility of capital markets. Table 2 displays a summary of analysis of Pearson correlation. A significant and positive relationship between Q and I (p < .01) indicates that firms increase investments as more options for investment appear. Consistent with previous studies, a positive and significant correlation exists between I and CF (p < .01), confirming that a firm’s investing activity is sensitive to cash flows. A negative and significant relationship between I and AL1 (p < .01) supports the argument of this study that lodging firms are more likely to reduce spending on capital investments as the level of asset-lightness increases. 4.2. Main analysis 4.2.1. Investment-cash flow sensitivities Table 3 shows the results from the estimations of Regression Model (4). To test Hypothesis 1, firms were categorized as financially constrained and unconstrained using three measures of financial constraints (KZ, SA, and WW Index). We used GMM estimators to cope with endogeniety while all estimations controlled for firm- and year-fixed Table 2 Summary of Pearson correlations. Variable N Ii,t Qi,t CFi,t AL1i, Ii,t Qi,t CFi,t AL1i,t ROICi,t SIZEi,t LEVi,t FSi,t LSi,t 1,478 1,478 1,478 1,478 1,478 1,478 1,478 1,478 1,478 1.00 0.22** 0.28** −0.17** 0.15** −0.21* −0.23* −0.14** 0.11** 1.00 0.85** 0.28* 0.83* −0.24* −0.49** −0.33* 0.17** 1.00 0.21** 0.84** −0.31** −0.38** −0.31* 0.26** 1.00 0.32** 0.09 −0.12** 0.12** -0.09* * Significant at 0.05. ** Significant at 0.01. 174 t ROICi,t SIZEi,t LEVi,t FSi,t LSi,t 1.00 −0.20** −0.46** −0.36** 0.21** 1.00 −0.23* −0.04 0.01 1.00 0.31* −0.28** 1.00 −0.86* 1.00 International Journal of Hospitality Management 78 (2019) 169–178 K. Seo, J. Soh Table 4 Investment-cash flow sensitivity and asset-light business model. Panel A (I) Qi,t (1) (2) VIF CFi,t * AL1i,t ** Unconst (2) SA Const Unconst (3) WW Const Unconst (CF AL1)i,t ** 0.033 (2.38) 0.033* (2.33) 0.054 (4.36) 0.048** (4.19) −0.097 (−3.22) −0.066** (−3.31) −0.878** (−3.76) 1.87 3.14 2.69 4.36 Panel B (1) KZ Const (II) * N J Qi,t 1,478 .23 1,478 .43 CFi,t * (CF*AL2)i,t AL2i,t ** ** 0.032 (2.37) 0.033* (2.25) 0.076 (4.54) 0.023** (4.83) −0.113 (−3.38) −0.089** (−3.38) −0.685** (−3.64) 1.96 3.66 3.75 5.41 (I) N J 1,478 .28 1,478 .20 (II) * Qi,t CFi,t AL1i,t (CF AL1)i,t N J Qi,t CFi,t AL2i,t (CF*AL2)i,t N J 0.036* (2.46) 0.007 (1.31) 0.044* (2.77) 0.044 (1.49) −0.138 (−1.87) −0.026* (−2.46) −0.734** (−3.20) −0.081 (−1.72) 444 .25 −0.221 (−1.49) −0.151* (−2.38) −0.971** (−3.36) −0.112 (−1.52) .32 .53 0.034** (3.55) 0.023 (1.33) 444 443 0.043* (2.27) 0.005 (1.39) 443 .21 −0.066 (−1.76) 0.003 (1.32) 0.051* (2.38) 0.036 (1.20) −0.133** (−3.88) −0.133 (−1.38) −0.693* (−2.47) −0.141 (−1.07) 444 .29 −0.355* (−2.43) −0.251 (−1.53) −0.869** (−3.84) −0.183 (−1.22) .28 .33 0.044* (2.25) 0.028 (1.02) 444 443 −0.090 (−1.54) 0.015 (1.17) 443 .36 0.053* (2.26) 0.016 (1.37) 0.076* (2.30) 0.031 (1.05) −0.213 (−1.41) −0.117** (−3.48) −0.912** (−3.89) −0.149 (−1.82) 444 .31 −0.336 (−1.48) −0.159* (−2.23) −0.719** (−4.10) −0.188 (−1.27) .27 .55 0.056* (2.56) 0.043 (1.18) 444 443 0.081 (1.50) 0.014 (1.53) 443 .36 * Significant at 0.05. To further investigate the effect of ALBM on ICFS, firms were placed into a financially constrained and unconstrained group. Hypothesis 2(b) states that the moderating effect of ALBM is greater for financially constrained firms than unconstrained firms. In column (I) of Panel B, the coefficients of the interaction terms for constrained firms are negative and statistically significant (t=-3.20, -2.47, and -3.89), indicating that ICFS decreases as the level of asset-lightness increases. On the other hand, the interaction terms are consistently insignificant for all of the estimations for unconstrained firms in rows (1)-(3), suggesting that investment becomes insensitive to changes in cash flows when firms have unlimited access to capital markets. These findings are consistent in column (II) of Panel B where the interaction of cash flow and asset-lightness attracts negative and statistically significant coefficients in all constrained firm estimations while they are also consistently higher than those of the unconstrained firms. Therefore, Hypothesis 2(b) receives support. The findings of this study suggest that ALBM allows lodging firms to mitigate the extent to which the availability of cash flows influences investments, especially when lodging firms experience constraints for external financing. More specifically, contrary to Almeida and Campello (2007), the positive effect of tangible assets on ICFS was not found among lodging firms. Given that ALBM involves little or no capital investments, the economic benefits of low capital requirements could more than offset the effect of tangible assets on investment for lodging firms. Therefore, when examining firms’ investment decisions in the context of the lodging industry, industry-specific structures for assets must be considered carefully. therefore, Hypothesis 3 gains support. Hypothesis 4 argues that lodging firms are likely to benefit more from implementing the asset-light strategy when operating primarily in the full-service segment than the limited-service segment. In Column (2) of Panel A and Panel B, the coefficient on the interaction of AL1 (AL2) and FS is statistically significant and is positively related to ROIC (t = 3.62 and 3.76). On the other hand, the coefficient on the interaction of AL1 (AL2) and LS is positive but insignificant, indicating lodging firms operating more limited-service hotels yield insignificant effects relative to the structure of asset-light ownership. Hence, Hypotheses 4(a) and 4(b) receive support. This study additionally checks the robustness of the results to address potential empirical biases in the regressions’ estimations. To avoid potential measurement problems when Tobin’s q proxies for investment opportunities, sales growth was used as an alternative measurement for investment opportunities. Chen and Chen (2012) argued that since sales growth reflects market demand, growth is an important factor for investing decisions. Measurement of sales growth is the difference of sales in time t and t + 1, divided by the sales during t. Although not tabulated, additional analyses provided qualitatively similar results, confirming that the findings of this study are robust and lack problems from measurements. 5. Conclusions The present study examines how a specific business model relative to asset ownership structures adopted by lodging firms affects their investment decisions and efficiencies of investments. Notably, ALBM positively influenced ICFS while increasing the efficiencies of investments measured by ROIC, especially when financially constrained. The findings of this study are informative for lodging investors and shareholders, by highlighting the significance of ALBM’s connection with corporate investment behavior. In addition, this paper provides an alternative approach to investigating investment behavior and ROIC, considering unique ownership structures of assets for lodging firms. To date, no empirical research explored the link among investment, ROIC, 4.2.2. Return on invested capital Table 5 presents the results of GMM estimations of Regression Models (6) and (7). Instrument validity in GMM was checked via Hensen’s J-statistics. VIFs were all in acceptable ranges, showing no sign of severe multicollinearity. Hypothesis 3 predicts that due to reduced capital investment and cost efficiency, ALBM will positively affect ROIC. The main effect of ABLM on ROIC is positive and significant (t = 3.44 and 3.67) in Column (1) of Panel A and Panel B, and 175 International Journal of Hospitality Management 78 (2019) 169–178 K. Seo, J. Soh Table 5 Asset-light business model and return on invested capital. Panel A AL1i,t FSi,t LSi,t (1) (2) VIF 2.13 −0.086*** (−8.10) −0.066** (−4.21) 0.038 (1.72) 0.212** (4.35) 0.161* (2.31) 0.578** (3.62) 0.154 (1.49) −0.078*** (−8.21) −0.081** (−3.46) 1,478 0.55 1,478 0.73 ** 0.048 (3.44) 0.173** (5.09) 0.118** (5.26) (AL1*FS)i,t * (AL1 LS)i,t SIZE,t LEVi,t N J Panel B AL2i,t FSi,t LSi,t (1) (2) ** −0.088*** (−8.11) −0.066** (−5.24) 0.062 (2.25) 0.286** (4.78) 0.118 (1.75) 0.612** (3.76) 0.203 (1.54) −0.086*** (−8.36) −0.067** (−5.27) 1,478 0.36 1,478 0.42 * (AL2 FS)i,t * (AL2 LS)i,t SIZE,t LEVi,t N J 2.55 4.71 4.89 1.68 1.76 VIF * 0.065 (3.67) 0.181** (5.11) 0.124* (2.56) 2.69 1.99 2.63 3.12 5.01 4.83 1.59 1.72 * Significant at 0.05. ** Significant at 0.01. *** Significant at 0.001. industries with distinctive structures for assets’ ownership. The current study also provides several important implications for owners, managers, and investors of lodging firms. First, the negative moderating effect of ALBM on the relationship between investment and cash flows implies substantial incentives for reducing ownership of assets among lodging firms that have concerns for underinvestment. Given the investment of a firm is sensitive to changes in cash flows, the firm is more likely to underreact to investment opportunities especially when cash flows are low (Hovakimian and Hovakimian, 2009). Minimizing capitalized commitments, however, asset-light lodging firms can effectively avoid potential risks for underinvestment even when internally generated cash flows are not sufficient to finance growth opportunities. It is of particular importance for a lodging operator to expand its operation and achieve economies of scale by increasing the number of branded, geographically diverse properties (Sohn et al., 2014). Therefore, lodging firms could consider adopting ALBM to effectively utilize third-party resources to drive the growth of their operations especially when seeking strategic opportunities in highly desirable and profitable locations. Second, the stronger effect of ALBM among financially constrained lodging firms further suggests that lodging operators could strategically reduce asset ownership and transition to the asset-light structure to assist continued expansion and sustain necessary investments, especially when costs and availability of external capital adversely affect funding. In general, facing challenges and difficulties in accessing external finance will prevent firms from fully exploiting investment opportunities (Kim et al., 2019). However, discontinuing expansions and and assets’ ownership structures in the lodging industry. This study establishes that a strategic view of assets’ structure will help improve a lodging firm’s ability to invest effectively and deliver high performance. 5.1. Contributions and implications The findings of this study contribute to the literature in several ways: First, this study adds to the literature by developing an alternative approach that allows us to identify the effect of financial constraints on firm investment. Since the novel study by Fazzari et al. (1988), ICFS has been a topic of frequent debate in contemporary corporate finance literature. However, what drives this relationship is still unclear in the extant studies while they only examine financial variables. By incorporating the variable that establishes a link between ICFS and real corporate decisions, the current study finds that a lodging firm’s strategic asset-light structure plays an important role in the firm’s investing activity. In particular, this study provides new empirical evidence that with limited capital investment, investment decisions of asset-light lodging firms are less susceptible to changes in cash flow. Second, unlike the previous study (Almeida and Campello, 2007), the current study shows that assets’ tangibility does not seem to be of significant importance for investment decisions among lodging firms. While the Almeida and Campello study only included manufacturing firms, analogous analysis for lodging firms did not find a significant effect from tangible assets on firms’ investing activities. This contradiction suggests that interpretation of the findings of Almeida and Campello study should entail some caution when applied to other 176 International Journal of Hospitality Management 78 (2019) 169–178 K. Seo, J. Soh segments. Finally, while the asset-light strategy has been widely accepted in the industry, it is still unclear why some lodging firms choose to be more (or less) asset-light than others despite their operational similarities. In particular, contrary to the alleged advantages of ALBM, recent studies provide inconsistent empirical evidence that ALBM has no significant impact on performance, suggesting that ALBM may not work for all lodging firms (Blal and Bianchi, 2019; Low et al., 2015). Some lodging firms may still prefer a conventional asset-ownership model over ALBM because it helps them establish more collateral for financing and achieve higher levels of financial and quality control based on streamlined organizational processes for communication and decision making (Brookes and Roper, 2012; Deng et al., 2017; Perrigot et al., 2009). Therefore, future research efforts could focus on investigating the performances of lodging firms that implement ALBM to various degrees to provide a more comprehensive understanding of whether or not ALBM is actually beneficial to financial performance in the long term. reducing property improvements due to unfavorable borrowing conditions would be detrimental for a hotel’s market share and operational results. For example, entering into franchise agreements with independent luxury resorts and hotels, Marriott extended its brand and market share even during the economic downturn when investment capital was scarce. Introduced in late 2009 and 2010, Autograph Collection and Edition brands, respectively, enabled Marriott to capture luxury market travelers without significant capital investments in new markets. Hence, the owners and operators of lodging firms especially when encountering financial constraints should consider establishing asset-light structures to capitalize on all potentially profitable investment and growth opportunities. Third, this study illuminates an important implication for lodging firms operating in the full-service segment: Advantageously use ALBM to develop and achieve competitive advantages. Significantly higher ROIC found among heavily asset-light lodging firms suggest that ALBM can help lodging firms concentrate on effective allocation and utilization of strategic resources to improve the operational efficiency of their organizations. In today’s highly competitive environments, lodging firms need to provide a unique value proposition that can help attract and retain guests. This is particularly critical in the full-service segment in which customers have higher expectations for lodging products and services (So et al., 2013). Therefore, lodging owners and operators should adopt ALBM to allow investments directed at developing and strengthening core competencies that enable delivery of unique value and high quality service to customers. Focusing investment initiatives which support strategic priorities will allow firms to not only market a unique value to their guests but also differentiate products and services from the competition. For instance, lodging firms can direct their investment into developing strong technology platforms that improve operating efficiency in distribution, reservations, property management, revenue management, procurement systems, and customer’s loyalty programs. In 2015, InterContinental Hotels Group invested in mobile technology platforms and increased, by 40 percent to $1.2 billion, mobile bookings via its direct channel (InterContinental Hotels Group, 2014). To attract the tech-savvy millennial generation, Hilton improved its direct digital booking systems, allowing digital check-in and check-out, and room selection options for guests reserving directly (Hilton Worldwide Inc, 2015). As such, lodging firms should seek an asset-light strategy to utilize resources cost effectively and invest in necessary development of core competencies. References Abel, A., Eberly, J., 2011. How Q and cash flow affect investment without frictions: an analytic explanation. Rev. Econ. Stud. 78, 1179–1200. Aiken, L.S., West, S.G., 1991. Multiple Regression: Testing and Interpreting Interactions. Sage, Newbury Park, London. Almeida, H., Campello, M., 2007. Financial constraints, asset tangibility, and corporate investment. Rev. Financ. Stud. 5, 1429–1460. Almeida, H., Campello, M., Weisbach, M., 2004. The cash flow sensitivity of cash. J. Finance 59, 1777–1804. Alti, A., 2003. How sensitive is investment to cash flow when financing is frictionless? J. Finance 58, 707–722. Arellano, M., Bond, S., 1991. Some tests of specification for panel data: Monte Carlo evidence and an application to employment equations. Rev. Econ. Stud. 58, 277–297. Baker, M., Stein, J., Wurgler, J., 2003. When does the market matter? Stock prices and the investment of equity-dependent firms. Q. J. Econ. 118 (3), 969–1005. Barclay, M.J., Smith Jr., C., 1995. The maturity structure of corporate debt. J. Finance 50, 609–631. Blal, I., Bianchi, G., 2019. The asset light model: a blind spot in hospitality research. Int. J. Hosp. Manag. 76, 39–42. Bowen, H.P., Wiersema, M.F., 1999. Matching method to paradigm in strategy research: limitations of cross-sectional analysis and some methodological alternatives. Strateg. Manage. J. 20 (7), 625–636. Brookes, M., Roper, A., 2012. Realizing plural-form benefits in international hotel chains. Tour. Manag. 33, 580–591. Brown, J., Petersen, B., 2009. Why has the investment-cash flow sensitivity declined so sharply? Rising R&D and equity market developments. J. Bank. Financ. 33, 971–984. Calomiris, C., Hubbard, R.G., 1995. Internal finance and firm-level investment: evidence from the undistributed profits tax of 1936-37. J. Bus. 68, 443–482. Campa, J.M., Kedia, S., 2002. Explaining the diversification discount. J. Finance 57 (4), 1731–1762. Chen, H., Chen, S., 2012. Investment-cash flow sensitivity cannot be a good measure of financial constraints: evidence from the time series. J. Financ. Econ. 103 (2), 393–410. Chu, R.K.S., Choi, T., 2000. An importance-performance analysis of hotel selection factors in the Hong Kong hotel industry: a comparison of business and leisure travellers. Tour. Manag. 21 (4), 363–377. Damodaran, A., 2007. Return on capital (ROC), return on invested capital (ROIC) and return on equity (ROE): Measurement and implications. SSRN Working Paper Series. https://doi.org/10.2139/ssrn.1105499. Deng, X., Ong, S.E., Quia, M., 2017. Real estate risk, corporate investment and financing choice. J. Real Estate Financ. Econ. 1–27. Dennis Jr, W.J., 2011. Entrepreneurship, small business and public policy levers. J. Small Bus. Manag. 49 (1), 92–106. Diamond, D.W., He, Z., 2014. The theory of debt maturity: the long and short of debt overhang. J. Finance 69 (2), 719–762. Enz, C.A., 2010. The Cornell School of Hotel Administration Handbook of Applied Hospitality Strategy. Sage, Los Angeles. Enz, C.A., Canina, L., Palacios-Marques, D., 2013. The Relationship Between New Venture Entry Mode and Firm Performance. http://scholarship.sha.cornell.edu/articles/612. Erickson, T., Whited, T., 2000. Measurement error and the relationship between investment and q. J. Polit. Econ. 108, 1027–1057. Espino-Rodríguez, T.F., Lai, P.C., 2014. Activity outsourcing and competitive strategy in the hotel industry: the moderator role of asset specificity. Int. J. Hosp. Manag. 42, 9–19. Fazzari, S., Petersen, B., 1993. Working capital and fixed investment: new evidence on financing constraints. Rand J. Econ. 24, 328–342. Fazzari, S.M., Hubbard, R.G., Petersen, B.C., 1988. Financing constraints and corporate investment. Brookings Pap. Econ. Act. 2, 141–195. Fazzari, S.M., Hubbard, R.G., Petersen, B.C., 2000. Investment-cash flow sensitivities are 5.2. Limitations and suggestions for future research Despite its relevant findings, the current study has several limitations. Firstly, while consensus considers the nature of the lodging industries to be homogeneous, globally, generalization of the results of this study to other regions and countries is problematic. Although the U.S. may very well represent the largest market for budget and economy hotels, the share of the budget and economy segment is small in comparison to that of European markets. Furthermore, large chainaffiliated lodging operators in Europe such as AccorHotels tend to own or lease more hotel properties than rivaling U.S. hotels. Future research can extend the current study to analyze the effect of the asset ownership structure across different segments and countries. Secondly, due to limited data availability, the sample of this study mainly consisted of large publicly traded lodging firms while excluding lesser well-known and small-scale firms such as independent hotels, suggesting lack of effective generalization. While this is a common form of sample-selection bias in financial research, future researchers could expand the focus to compare the investment behavior and return on invested capital among various types of lodging firms. Thirdly, although the current study examined the asset-light business model in the full-service and limited-service segments, classifications of lodging firms at a more granular level could provide a better understanding of the asset-light strategies adopted by various lodging firms throughout various market 177 International Journal of Hospitality Management 78 (2019) 169–178 K. Seo, J. Soh useful: a comment on Kaplan and Zingales. Q. J. Econ. 115, 695–705. Gala, V.D., Gomes, J., 2013. Beyond Q: Estimating Investment without Asset Prices. SSRN Working Paper Series. id=2024716. http://papers.ssrn.com/sol3/papers.cfm? abstract. Gomes, J.F., 2001. Financing investment. Am. Econ. Rev. 91, 1263–1285. Gonzalez, R., Llopis, J., Gasco, J., 2013. Outsourcing and strategy in Spanish town halls: a field study. Manage. Decis. 51 (1), 97–119. Guedes, J., Opler, T., 1996. The determinants of the maturity of corporate debt issues. J. Finance 51 (5), 1809–1833. Hadlock, C., Pierce, J., 2010. New evidence on measuring financial constraints: moving beyond the KZ index. Rev. Financ. Stud. 23 (5), 1909–1940. Hansen, L.P., 1982. Large sample properties of generalized method of moments estimatiors. Econometrica 50, 1029–1054. Hansen, H., Huhn, W., Legrand, O., Steiners, D., Vahlenkamp, T., 2009. CAPEX Excellence: Optimizing Fixed Asset Investments. John Wiley and Sons. Hilton Worldwide Inc, 2015. Annual Report of Hilton Worldwide Holdings Inc. . http:// s1.q4cdn.com/246698567/files/annual-reports/2014/index.html. Himmelberg, C., Petersen, B., 1994. R&D and internal finance: a panel study of small firms in high-tech industries. Rev. Econ. Stat. 76, 38–51. Hoesli, M., Oikarinen, E., 2012. Are REITs real estate? Evidence from international sector level data. J. Int. Money Finance 31 (7), 1823–1850. Hovakimian, A., Hovakimian, G., 2009. Cash flow sensitivity of investment. Eur. Financ. Manag. 15 (1), 47–65. Hovakimian, G., Titman, S., 2006. Corporate investment with financial constraints: sensitivity of investment to funds from voluntary asset sales. J. Money Credit Bank. 38 (2), 357–374. Hyatt Hotels Corp, 2017. Annual Report of Hyatt Hotels Corporation. . http://s2.q4cdn. com/278413729/files/doc_financials/annual_2017/Q4-2017-Form-10-K.pdf. InterContinental Hotels Group, 2014. Annual Report of InterContinental Hotels Group. . http://www.ihgplc.com/files/reports/ar2014/files/IHG_Report_2014.pdf. Jiang, B., Frazier, G.V., Prater, E.L., 2006. Outsourcing effects on firms’ operational performance: an empirical study. Int. J. Oper. Prod. Manage. 26 (12), 1280–1300. Johnson, S.A., 2003. Debt maturity and the effects of growth opportunities, managerial discretion, and the security issue decision. Rev. Financ. Stud. 16, 209–236. Kaplan, S.N., Zingales, L., 1997. Do investment-cash flow sensitivities provide useful measures of financing constraints? Q. J. Econ. 112, 169–215. Kaplan, S.N., Zingales, L., 2000. Investment-cash flow sensitivities are not valid measures of financing constraints. Q. J. Econ. 115, 707–712. Khramov, M.V., 2012. Asymmetric Effects of the Financial Crisis: Collateral-Based Investment-cash Flow Sensitivity Analysis (No. 12-97). International Monetary Fund. Kim, H., Gu, Z., 2004. Financial determinants of corporate bond ratings: an examination of hotel and casino firms. J. Hosp. Tour. Res. 28 (1), 95–108. Kim, J., Jang, S., 2012. Do hotel REIT companies face investment constraints? A comparison with C-corporation hotel companies. Int. J. Hosp. Manag. 31 (2), 573–578. Kim, W., Cho, M., Brymer, R.A., 2013. Determinants affecting comprehensive propertylevel hotel performance: the moderating role of hotel type. Int. J. Hosp. Manag. 34, 404–412. Kim, S.H., Noh, S., Lee, S.K., 2019. Asset-light strategy and real estate risk of lodging Ccorps and REITs. Int. J. Hosp. Manag. https://doi.org/10.1016/j.ijhm.2018.09.004. Lai, J., Lui, S.S., Hon, A.H.Y., 2014. Does standardized service fit all? Int. J. Contemp. Hosp. Manage. 26 (8), 1341–1363. Lee, K.J., 2014. Attitudinal dimensions of professionalism and service quality efficacy of frontline employees in hotels. Int. J. Hosp. Manag. 41, 140–148. Lee, S.K., 2015. Quality differentiation and conditional spatial price competition among hotels. Tour. Manag. 46, 114–122. Lewellen, J., Lewellen, K., 2016. Investment and cash flow: new evidence. J. Financ. Quant. Anal. 51 (4), 1135–1164. Li, K., Tang, V., 2008. Earnings quality and future capital investment: evidence from discretionary accruals. Georgetown University Working Paper. Liou, F., 2011. The effects of asset-light strategy on competitive advantage in the telephone communications industry. Technol. Anal. Strateg. Manag. 23 (9), 951–967. Low, W., Das, P., Piffaretti, C., 2015. The role of hotels in mixed asset portfolios: revisiting the asset-heavy versus asset-light debate. Int. J. Built Environ. Asset Manag. 1 (4), 273–292. Maly, J., Palter, R.N., 2002. Restating the value of capital light. McKinsey on Finance 5, 1–5. Moyen, N., 2004. Investment-cash flow sensitivities: constrained vs. unconstrained firms. J. Finance 59, 2061–2092. Myers, S.C., Majluf, N.S., 1984. Corporate financing and investment decisions when firms have information that investors do not have. J. Financ. Econ. 13 (2), 187–221. Page, T., 2007. ‘Asset-light’- managing or leasing? J. Retail Leisure Property 6 (2), 97–99. Peng, J., Zhao, X., Mattila, A., 2015. Improving service management in budget hotels. Int. J. Hosp. Manage. 49, 139–148. Perrigot, R., Cliquet, G., Piot-Lepetit, I., 2009. Plural form chain and efficiency insights from the French hotel chains and the DEA methodology. Eur. Manage. J. 27, 268–280. Prahalad, C.K., Hamel, G., 1990. The core competence of the corporation. Harvard Bus. Rev. 71, 79–91. Roh, Y.S., 2002. Size, growth rate and risk sharing as the determinants of propensity to franchise in chain restaurants. Int. J. Hosp. Manage. 21, 43–56. Rothaermel, F.T., Hess, A.M., 2007. Building dynamic capabilities: innovation driven by individual, firm, and network-level effects. Organiz. Sci. 18, 898–921. Silva, R., 2015. Multimarket contract, differentiation, and prices of chain hotels. Tour. Manage. 48, 305–315. Simon, J.B., 2015. The Capital-Light Continuum. http://www.hotelnewsnow.com/ Article. So, K.K.F., King, C., Sparks, B.A., Wang, Y., 2013. The influence of customer brand identification on hotel brand evaluation and loyalty development. Int. J. Hosp. Manage. 34, 31–41. Sohn, J., Tang, C., Jang, S., 2013. Does the asset-light and fee-oriented strategy create value? Int. J. Hosp. Manage. 32, 270–277. Sohn, J., Tang, C., Jang, S., 2014. Asymmetric impacts of the asset-light and fee-oriented strategy: the business cycle matters!. Int. J. Hosp. Manage. 40, 100–108. Stohs, M.H., Mauer, D.C., 1996. The determinants of corporate debt maturity structure. J. Bus. 69 (3), 279–312. Tabachnick, B.G., Fidell, L.S., 1996. Using Multivariate Statistics. HarperCollins College Publishers, New York. Tanford, S., Raab, C., Kim, Y.S., 2011. The influence of reward program membership and commitment on hotel loyalty. J. Hosp. Tour. Res. 35 (3), 279–307. Tanford, S., Raab, C., Kim, Y.S., 2012. Determinants of customer loyalty and purchasing behavior for full-service and limited-service hotels. Int. J. Hosp. Manage. 31, 319–328. Tang, Y.C., Liou, F.M., 2010. Does firm performance reveal its own causes? the role of Bayesian inference. Strategic Manage. J. 31 (1), 39–57. Thompson, S.B., 2011. Simple formulas for standard errors that cluster by both firm and time. J. Financial Econ. 99, 1–10. Tuzel, S., 2010. Corporate real estate holdings and the cross-section of stock returns. Rev. Financ. Stud. 23 (6), 2268–2302. Upneja, A., Sharma, A., 2009. Relationship between changes in cash flow and investments in publicly traded restaurant firms in the United States. Hospitality Rev. 27 (2), 61–80. Whited, T.M., Wu, G., 2006. Financial constraints risk. Rev. Financ. Stud. 19 (2), 531–559. Wooldridge, J.M., 2002. Econometric Analysis of Cross Section and Panel Data. MIT Press, Cambridge. Wu, C.H., Liao, H.C., Hung, K.P., Ho, Y.H., 2012. Service guarantees in the hotel industry: their effects on consumer risk and service quality perceptions. Int. J. Hospitality Manage. 31, 757–763. Younes, E., Kett, R., 2007. Hotel investment risk: What are the chances? J. Retail Leisure Property 6 (1), 69–78. Yu, S.M., Liow, K.H., 2009. Do retail firms benefit from real estate ownership? J. Property Res. 26 (1), 25–60. Zhang, H.Q., Ren, L., Huawen, S., Xiao, Q., 2013. What contributes to the success of home inns in China? Int. J. Hosp. Manage. 33, 425–434. 178