Liquidity and speed of leverage adjustment

Abstract

This article investigates the effect of liquidity on the speed of adjustment (SOA) of corporate leverage at the individual company level. Using panel analysis of data from 35 countries between 1996 and 2016, we find that high-liquidity firms have a significantly faster SOA than less liquid firms. This result survives a series of robustness checks and holds after addressing the endogeneity concern using exogenous shocks and additional control variables. We find that the positive effect of liquidity on the SOA exists only for over-levered firms, and this impact is moderated in countries with bankruptcy codes. We further find that the positive liquidity-SOA relationship is less (more) pronounced for firms in strong (weak) institutional environments. The results provide new insights into the role of liquidity in firms’ capital structure decisions and the determinants of capital structure dynamics.

JEL Classification: G12, G32

Keywords

Dynamic trade-off theory institutional environments leverage SOA liquidity

1. Introduction

The static trade-off theory of capital structure predicts that a firm can maximize its value by operating at a target debt ratio that balances the benefits and costs of debt financing. The dynamic trade-off view suggests that when firms deviate from their target ratios, they will make adjustments to move back to the target (Fischer et al., 1989; Goldstein et al., 2001; Leary and Roberts, 2005). However, due to financing frictions, this adjustment can involve nontrivial costs, making the capital structure adjustment dynamic in nature and the speed of adjustment (SOA) unexpectedly slow.

In recent years, there has been growing interest in the critical role of liquidity in capital structure decisions. According to the static trade-off model, more liquid firms have lower floatation costs for equity issuance, which makes equity financing more attractive than debt financing. As a result, high-liquidity firms are likely to have lower leverage. Lipson and Mortal (2009) and Udomsirikul et al. (2011) show the negative impact of liquidity on capital structure in the US and Thailand markets, respectively. Using a global dataset, Gao and Zhu (2015) document that high-liquidity firms are expected to have lower debt financing in their capital structure and that this relationship is more pronounced in countries with weak institutional environments. Exploring the Australian context, Nadarajah et al. (2018) suggest a significantly negative liquidity-leverage relationship and find that high-liquidity firms have significantly negative corporate governance-leverage relationships whereas low-liquidity firms do not have this association. These studies, however, focus on the static trade-off view of capital structure. It is, therefore, interesting to know how liquidity affects the dynamic nature of capital structure, specifically the speed at which firms adjust their capital structure toward the target. Furthermore, given the increasingly important role of institutional environments in firms’ financial policies (Bancel and Mittoo, 2004; Demirgüç-Kunt and Maksimovic, 1999; Öztekin, 2015; Öztekin and Flannery, 2012), we are motivated to investigate the liquidity-leverage SOA relationship, in conjunction with the impacts of a country’s institutional environments.

Theories of corporate finance have highlighted the important role of liquidity in reducing transaction costs (Amihud and Mendelson, 1986; Butler et al., 2005; Dang et al., 2015; Lipson and Mortal, 2009). An issuer of either debt or equity will have to pay additional costs to underwriters/intermediaries (such as investment banks and financial institutions) when raising more capital. Butler et al. (2005) find that underwriters charge higher fees when they assist illiquid firms in the issuance process. In addition, lower transaction costs can come from improvement in the corporate governance of high-liquidity firms (Edmans et al., 2013; Maug, 1998; Noe, 2002). Liquidity can facilitate the exercise of governance activities because it allows large shareholders to engage in correcting managerial failures and covering monitoring costs through informed trading. As a result, it reduces information asymmetry, agency costs, and the costs of leverage adjustment (Chang et al., 2014; Liao et al., 2015). The bottom line is that a liquid firm will face lower costs associated with issuing new securities and thus be more inclined to rapidly correct any deviation of its actual leverage ratio from its target ratio.

Based on the above arguments, we expect that a firm’s liquidity has an impact on the leverage SOA to the target leverage ratio. To date, however, this important link between firm-level liquidity and dynamic capital structure has been overlooked in the literature. In this article, we fill this gap and examine the hypothesis that a firm’s liquidity tends to increase its leverage SOA. In addition, the pecking order theory suggests that the leverage SOA is asymmetric as over-levered firms are faced with more costly funding channels than under-levered firms (Byoun, 2008). We argue that liquidity decreases the adjustment costs of equity financing and may affect over-levered firms rather than under-levered firms (Butler et al., 2005). Accordingly, we empirically examine whether liquidity has a distinct impact on leverage SOA in over- and under-levered firms.

Furthermore, at the country level, we investigate how institutional strength affects the relationship between firm-level liquidity and leverage SOA. Institutional environments are widely perceived as external control mechanisms that reduce agency conflicts and serve as a “cheap” form of macro-level investor protection. For firms that operate in countries with strong legal and political institutions, the monitoring costs of large shareholders could be lower than those of firms operating in countries with weak institutional strength. Moreover, in countries with a strong institutional environment, firm-level information tends to be released to the market in a more accurate and timely manner, which lowers the firm’s external financing costs and accelerates the leverage SOA (Öztekin and Flannery, 2012). Strong institutional settings and liquidity, therefore, can substitute each other in improving a firm’s leverage SOA by reducing the adjustment costs. As an alternative argument, strong institutional environments can also enhance the capital market efficiency, provide better enforcement mechanisms and more transparent information, and ensure investor and creditor protection (An et al., 2015). Firms operating in these countries can enjoy less asymmetric information and have higher liquidity. In general, strong institutional environments and liquidity can complement each other in reducing the costs of adjusting corporate leverage.

To test our hypotheses, we use an international sample of 35 countries from 1996 to 2016. Our empirical results are consistent with our hypotheses and suggest three key findings. First, firm-level liquidity has a significant and positive effect on the SOA to the target leverage ratio. We find that high firm-level liquidity not only reduces equity financing costs but also improves corporate governance, which results in lower costs of leverage adjustment. This finding is consistent with the previous literature on microfinance (Amihud and Mendelson, 1986; Chang et al., 2014; Dang et al., 2015; Liao et al., 2015; Lipson and Mortal, 2009; Maug, 1998; Noe, 2002). Second, we find that liquidity only has a significantly positive impact on the leverage SOA of over-levered firms, and this impact is moderated in countries that have bankruptcy codes. For under-levered firms, however, the association is ambiguous. Third, we find that strong institutional environments, as proxied by the strength of law and order, risk of expropriation, risk of contract repudiation by government, level of corruption (La Porta et al., 1998), creditor rights enforcement (Djankov et al., 2003), and the significance of the banking sector (Demirgüç-Kunt and Maksimovic, 1996), tend to attenuate the positive association between liquidity and leverage SOA.

Our results contribute to the current literature in the following aspects. First, by using dynamic partial adjustment models of capital structure to analyze the determinants of SOA, this article introduces new evidence of the impact of firm-level liquidity on SOA. Consistent with the extant empirical literature on dynamic capital structure adjustments, we contribute to the growing literature on the determinants of the leverage SOA (An et al., 2015; Çolak et al., 2018; Devos et al., 2017; Faulkender et al., 2012; Öztekin and Flannery, 2012). In particular, this article adds to recent research on the relationship between the sensitivity of cost of equity on leverage deviation and leverage SOA (Zhou et al., 2016). Our article shows that firm-level liquidity affects both capital structure (Lipson and Mortal, 2009) and the SOA of leverage, and this effect is vastly different between over- and under-levered firms.

Second, we contribute to the current literature on the importance that firms place on the target leverage level by investigating the distinct impacts of liquidity on the leverage SOA for over- and under-levered firms. Byoun (2008) shows that firms have differential adjustment speeds conditional on whether the actual leverage ratios are below or above the targets. Warr et al. (2012) suggest that the impact of equity mispricing on the speed of leverage adjustment depends on whether the firm is above or below its target leverage. An et al. (2015) also conclude that the position of the target leverage ratio affects the relationship between crash-risk exposure and corporate leverage SOA. In this article, we add to this strand of the literature by showing that the impact of liquidity on leverage SOA is asymmetric and depends on whether the firm is over or under its target leverage ratio.

Third, by using an international sample, we investigate how country-level environments affect the sensitivity of leverage SOA to liquidity, thus contributing to the extant literature on the effects of macro-level institutional environments on corporate capital structure decisions and aggregate financial markets. For instance, Öztekin and Flannery (2012) find evidence that firms that operate in countries with better institutional settings have lower external financing costs and higher leverage SOA. Çolak et al. (2018) conclude that high-quality institutions attenuate the adverse effects of uncertainty on leverage adjustment. Shleifer and Wolfenzon (2002) and La Porta et al. (2002) demonstrate that liability enforcement and strict legal disclosure requirements benefit financial market developments, establishing the links between institutional environments and the aggregate financial markets.

The article proceeds as follows. Section 2 reviews the related literature and develops the hypotheses. Section 3 presents the data and constructs the variables in our empirical study. Section 4 explains the empirical methods. Section 5 presents the empirical results, including the descriptive statistics, correlation analysis, baseline results, and the endogeneity correction. Section 6 concludes the article.¹

2. Related literature and hypotheses development

2.1. Literature on leverage adjustment

Our hypotheses are developed based on the literature relating to dynamic leverage adjustment. This strand of literature focuses on the adjustment of firms’ leverage ratios toward their target levels, and specifically, the determinants of leverage SOA.

Following the fundamental framework of Modigliani and Miller (1958), multiple studies have emphasized the trade-off theory—a principal view of capital structure (Fischer et al., 1989; Goldstein et al., 2001; Strebulaev, 2007). This body of research has shown that firms have a time-varying target leverage at which various costs (e.g. agency costs due to the conflict between debtholders and stockholders, and bankruptcy costs or financial distress costs) and benefits (e.g. tax savings, and mitigating agency costs arising from the conflict between stockholders and managers) of debt are optimally balanced. Existing empirical studies also support this view that firms have target leverages and attempt to move toward these targets (Byoun, 2008; Flannery and Rangan, 2006; Huang and Ritter, 2009).

However, the process of adjusting to the target levels is costly. The literature on capital structure shows that the speed at which firms converge to their targets varies, with each firm facing different adjustment costs. These costs can include issuance costs and/or opportunity costs. For instance, Strebulaev (2007) and Goldstein et al. (2001) find that firms with lower transaction costs adjust their leverage more frequently. Faulkender et al. (2012) suggest that the variation in speed of leverage adjustments is due to the differences in sunk and incremental costs. This has motivated a line of research into the cross-sectional variations in the SOA and conditions on specific opportunity costs that affect the SOA. In particular, Drobetz and Wanzenried (2006) argue that faster-growing firms and those that diverge further away from their optimal leverage will adjust their capital structure positions more quickly. Chang et al. (2014) and Liao et al. (2015) provide evidence for the association between corporate governance and dynamic capital structure. An et al. (2015) show that a firm’s crash-risk exposure has a significantly negative impact on leverage SOA. Zhou et al. (2016) and Devos et al. (2017) examine the impact of the cost of equity and debt covenants, respectively, on leverage SOA. Most recently, Dang et al. (2019) show that firms with greater news coverage and more positive news sentiment have greater leverage adjustment speeds. In general, these studies imply that firms adjust faster to their target leverage when the adjustment costs are low.

Another strand of research shows that market timing as equity mispricing has a significant impact on the costs of leverage adjustment (Flannery and Rangan, 2006; Liu, 2009; Warr et al., 2012). When a firm’s equity is overvalued, it is cheaper for the firm to adjust its leverage ratio through equity instead of debt. Flannery and Rangan (2006) consider the effect of market timing that is proxied by a firm’s market-to-book ratio on leverage adjustment models. Liu (2009) shows that historical market-to-book ratios as a proxy for stock market misevaluation have a significant effect on leverage ratio even when firms do not attempt to time the market. More recently, Warr et al. (2012) find that equity mispricing affects a firm’s SOA, although this impact depends on whether the firm is under- or over-levered.

The effects of macroeconomic conditions, business cycles, institutional factors, and political uncertainty on the SOA have also been investigated (Çolak et al., 2018; Cook and Tang, 2010; Drobetz et al., 2015; Öztekin and Flannery, 2012). Hackbarth et al. (2006) show that if firms make financial decisions based on the business cycle, macroeconomic conditions should have a significant impact on leverage adjustments. Cook and Tang (2010) confirm this inference by demonstrating that firms move faster to the targets in good macroeconomic states compared to bad states. Elsas and Florysiak (2011) provide empirical evidence of firms adjusting their leverage faster in countries with low expected bankruptcy costs and low default risks. In analyses of a wide range of institutional determinants, Öztekin and Flannery (2012) and Öztekin (2015) also confirm that good legal and institutional environments lower a firm’s leverage adjustment costs. Most recently, Çolak et al. (2018) illustrate that political and economic uncertainty dramatically slows down the speed of a firm’s leverage adjustments.

2.2. Liquidity in determining leverage SOA

The important role of liquidity in capital structure transaction costs is well documented in the literature. Stoll and Whaley (1983) note that when evaluating equity investments, investors should consider transaction costs as an important element. This type of cost, they argue, may explain the higher required rates of return on illiquid stocks. Amihud and Mendelson (1986) also propose that due to high transaction costs, such as tax, illiquid equity has high required rates of return. Similarly, Butler et al. (2005) concentrate on issuance costs and find that investment banks charge lower fees for high-liquidity firms. In addition, liquidity is also found to improve the corporate governance that is considered by underwriters and thus lead to a lower fee for equity issuance (Edmans et al., 2013; Kahn and Winton, 1998; Maug, 1998; Noe, 2002). Liquidity allows large shareholders to monitor corporate management and cover monitoring costs through informed trading, thus enabling the exercise of governance activities. To the extent that liquidity can enhance corporate governance, it may also reduce information asymmetry, agency costs, and therefore the cost of leverage adjustment (Chang et al., 2014; Liao et al., 2015). In general, this body of research suggests that stocks with higher liquidity have lower transaction costs, which are among the main elements of leverage adjustment costs.

Brennan and Subrahmanyam (1996) and Brennan et al. (1998) also provide empirical evidence to support a negative relationship between liquidity and cost of equity. However, there is a strand of literature that establishes a relationship between cost of equity and leverage SOA. For instance, Öztekin and Flannery (2012) argue that firms with low trading costs enjoy significantly higher estimated adjustment speeds. Zhou et al. (2016) report an indirect link between the cost of equity and leverage SOA. Combining these two strands of literature together may address the implications of equity liquidity for leverage SOA.

An alternative explanation of the relationship between liquidity and the SOA in capital structures can be drawn from the pecking order theory introduced by Myers and Majluf (1984). This theory argues that adverse selection problems will be smaller for firms with higher liquidity. Consequently, such firms face lower adverse costs and thus lower transaction costs, which will translate to a faster SOA.

Based on the above discussion, we propose our main hypothesis as:

H1. Liquidity has a positive and significant impact on leverage SOA.

2.3. The distinction between over-levered and under-levered firms

To investigate the importance that firms place on a target leverage ratio, we separately examine the relationship between liquidity and the leverage SOA for over- and under-levered firms. To adjust to the target ratio, an over-levered firm (i.e. leveraged above its target levels) usually needs to substitute equity for debt. Therefore, for over-levered firms, those with high liquidity have low costs for access to equity capital and thus adjust their leverage at relatively low marginal costs (Frieder and Martell, 2006; Lipson and Mortal, 2009). Consequently, the SOA toward their target levels is likely to be faster for those with high liquidity than for their low-liquidity counterparts.

On the other hand, an under-levered firm (i.e. leveraged below its target levels) needs to substitute debt for equity to adjust to the target. As suggested by Lipson and Mortal (2009), firms with high liquidity prefer equity financing when raising capital. These firms have a lower cost of equity and greater reliance on equity financing. If the firm’s investment opportunities are so substantial that funding them requires external capital, the firm’s managers may consider that the benefits of increasing the firm’s capital by increasing equity and deviating further from targets outweigh the value enhancement by moving closer to their targets. Conversely, if firms have few investment opportunities, under-levered high-liquidity firms face low costs of accessing the equity capital market but may rarely do so as they consider the benefits of adjusting to the targets are higher than the benefits of increasing the firm’s equity capital. Consequently, liquidity is less likely to affect the leverage SOA of under-levered firms.

From the above discussion, we propose the next hypothesis:

H2. The impact of liquidity on the SOA is more (less) pronounced for over-levered (under-levered) firms.

2.4. The role of institutional environments

Institutional environments are generally perceived as external governance mechanisms to reduce agency conflicts and as “low-cost” forms of macro-level investor protection. They are commonly set beyond the control of firms and are less costly control mechanisms than internal ones. The literature shows that countries with better institutions have more information transparency and lower transaction costs. Furthermore, in countries with strong institutional environments, firm-level information tends to be released to the market in a more accurate and timely manner (An et al., 2015; Çolak et al., 2018; Öztekin, 2015; Öztekin and Flannery, 2012). For instance, Doidge et al. (2007) find that country- and economic-level characteristics can explain transparency much better than firm-level characteristics. Öztekin and Flannery (2012) and Öztekin (2015) show that the transaction cost of external financing is lower and the SOA is higher in countries with better information environments. Çolak et al. (2018) conclude that high-quality institutions can be redesigned to offset the adverse impacts of uncertainty on leverage SOA. Thereby, we argue that due to the greater information transparency, there is a more limited role for price information revelation through trading, and thus a more limited role for liquidity in countries with good institutions. On the other hand, in countries with poorer institutions and less transparent information, stock market liquidity facilitates price information revelation through trading, and hence lower transaction costs, implying a stronger relationship between liquidity and leverage SOA. Therefore, strong institutional settings and liquidity may substitute each other in improving a firm’s leverage SOA by reducing the adjustment costs. Thus, our third hypothesis is proposed as:

H3a. The impact of liquidity on the SOA is less (more) pronounced in countries with strong (weak) institutional environments.

As an alternative argument, strong institutional environments and liquidity can complement each other in reducing the costs of adjusting firms’ leverage. Since strong institutional environments enhance capital market efficiency, provide better enforcement mechanisms and more transparent information, and ensure investor and creditor protection, firms operating in these countries can enjoy lower transaction costs, less asymmetric information, and higher liquidity. Brockman and Chung (2003) show that a good institutional environment diminishes information asymmetries and decreases the probability of trading against informed traders, which leads to lower bid-ask spread and better liquidity. Doidge et al. (2007) suggest that country-level factors have much higher explanatory power for governance quality and transparency of firms compared to firm-level factors. Thus, an alternative hypothesis to H3a is as follows:

H3b. The impact of liquidity on the SOA is more (less) pronounced in countries with strong (weak) institutional environments.

3. Data and variable definitions

3.1. Data

We obtain annual firm-level and industry-level accounting data from Worldscope and daily trading data (e.g. share prices, stock returns, and trading volumes) from Datastream for publicly traded firms around the world. We also retrieve information about macro-level institutional environments from La Porta et al. (1998), Djankov et al. (2003), and Djankov et al. (2008). Country-level data for control variables are obtained from the World Development Indicators (WDI) and OECD databases. Only firms with common securities are included and those with special features, such as warrants, trusts, funds, and non-equity stocks, are excluded. We also disregard financial and utility corporations since these corporations are subject to special regulations on financing policies. Our initial sample consists of 763,131 observations. To reduce short panel bias, we require the firms to have data for all variables for at least two consecutive years (552,168 observations). We also remove observations with leverage ratios beyond the unit interval (18,839 observations). Following Çolak et al. (2018), we exclude countries that have fewer than 10 firms with available accounting data to ensure reasonable cross-sectional variations within a country (1371 observations). We winsorize all the continuous variables at the 1st and 99th percentiles to mitigate the potential impact of extreme values. After applying these filters, our final sample consists of 190,754 firm-year observations for 16,963 unique firms in 35 countries over the period spanning 1996 to 2016.²

3.2. Variable construction

3.2.1. Leverage

We measure our dependent variable, leverage, using both the book ratio $(B L E V)$ and the market ratio $(M L E V)$ following existing studies (An et al., 2015; Chang et al., 2014; Flannery and Rangan, 2006; Öztekin and Flannery, 2012). Book leverage is measured as the ratio of the book value of total debt and the book value of total assets. Market leverage is calculated as the book value of total debt divided by the sum of the market value of equity and the book value of total debt.

3.2.2. Liquidity

Following the literature in liquidity, we use three proxies for liquidity: Amihud illiquidity (Amihud, 2002; Lipson and Mortal, 2009; Nadarajah et al., 2018), Turnover (Berkman and Nguyen, 2010; Goyenko et al., 2009; Lipson and Mortal, 2009), and High Low impact (Corwin and Schultz, 2012; Fong et al., 2017). These are the most popular measures of liquidity in major analyses, especially for international data.

Specifically, the Amihud (2002) illiquidity measure is defined as the average ratio of the daily absolute stock return divided by the dollar value of volume. This ratio reflects the daily price change related to one dollar of trading volume, or the daily price impact of the order flow. In this study, we use the annual average of this daily liquidity measure for each stock i

A m i h u d_{i, t, j} = 1 / D_{i, t, j} \sum_{1}^{D_{i, t}} \frac{| R_{i, t, d} |}{D V O L_{i, t, d}}

where $D_{i, t}$ is the number of days for which the volume of stock i in year t. The Amihud illiquidity measure represents the price movement associated with trading volume or the price impact of the order flow. Stocks are considered as illiquid if small volume trades lead to high price movement. In relation to different price impact measures, Hasbrouck (2009) shows that Amihud illiquidity is the strongest measure that reflects the price impact of stocks. To improve the interpretation of the results, we multiply the Amihud illiquidity measure by (–1) to convert it into a liquidity proxy. We also multiply the ratio by $10^{5}$ as in the standard liquidity literature.

Second, stock turnover $(T u r n_{i, t, j})$ , which has been used as a general measure of liquidity in a large number of previous studies (Berkman and Nguyen, 2010; Datar et al., 1998) is defined as the number of shares traded in a day, divided by the total number of shares outstanding

T u r n_{i, t, j} = \frac{D V O L_{i, t, j}}{s h a r e_{i, t, j}}

where $V O L_{i, t, j}$ is the total trading volume for stock i in year t, in country j and $s h a r e_{i, t, j}$ is the number of shares outstanding for stock i in year t, and country j. Volume data for each stock is obtained on a daily basis, while data on the number of shares outstanding are obtained on a yearly basis.³ The use of turnover rate as a measure of liquidity has strong theoretical support (Chai et al., 2010; Chan and Faff, 2003), especially in relation to turnover denoting the average holding period of stocks. The smaller the turnover ratio, the longer is the average holding period. Based on the work of Amihud and Mendelson (1986), stocks with higher spread have comparatively longer expected holding periods. Therefore, turnover should be negatively associated with spread and positively associated with liquidity.

Third, we use the high-low impact estimator introduced by Corwin and Schultz (2012) as a proxy for capturing high-frequency liquidity benchmarks on a global basis (Fong et al., 2017).⁴ For a given firm i, the high-low impact is measured as

H i g h L o w_{i, t, j} = \frac{2 (\exp (α) - 1)}{\exp (α) + 1}

where

α = \frac{\sqrt{2 β} - \sqrt{β}}{3 - 2 \sqrt{2}} - \sqrt{\frac{γ}{3 - 2 \sqrt{2}}}

β = {[\log (\frac{H_{t}}{L_{t}})]}^{2} + {[\log (\frac{H_{t + 1}}{L_{t + 1}})]}^{2}

γ = {[\log (\frac{\max {H_{t}, H_{t + 1}}}{\max {L_{t}, L_{t + 1}}})]}^{2}

Assuming that during day t, the highest trading price $H_{t}$ and the lowest trading price $L_{t}$ are from buy and sell trades, respectively. The high-low ratio reflects the information asymmetry of the two trading sides. A high High-Low estimator implies high liquidity and a low discrepancy between the highest and lowest trading prices across a 2-day period.

3.2.3. Country-level variables

Next, in order to study how institutional environments impact the relationship between liquidity and leverage SOA, following previous studies (An et al., 2015; Gao and Zhu, 2015; Öztekin, 2015; Öztekin and Flannery, 2012), we consider several macro-level variables from La Porta et al. (1998), Djankov et al. (2003), and Gao and Zhu (2015). Specifically, La Porta et al. (1998) suggest that the strength of law and order can protect investors against expropriation by management and ensure the implementation of investors’ rights when necessary, thereby lowering firms’ leverage adjustment costs. Consequently, we account for law and order, RulLaw, which measures the law and order tradition in the country. Good governments also respect property rights, encourage informed trading on stock markets, and are associated with better investor protections, leading to reduced costs of adjusting to target leverage and increased SOA (Morck et al., 2000). Following La Porta et al. (1998), we define good governments in terms of the risk of expropriation (RisExp), risk of contract repudiation by government (RisCon), and level of corruption (Corrup). Furthermore, Rajan and Zingales (1995) show that strong creditor rights enforcement enhances ex ante contractibility, making it easier to access the debt market, lower a firm’s rebalance costs, and speed up firms’ leverage SOA. To measure the enforcement of debt contract, following Öztekin and Flannery (2012) and Öztekin (2015), we use creditor rights enforcement (CreEnf), which measures substantive and procedural statutory intervention in judicial cases (Djankov et al., 2003). Furthermore, a sophisticated banking system helps firms to access external finance, increasing the leverage adjustment speed (Gao and Zhu, 2015). We use the ratio of domestic credit provided by the banking sector to gross domestic product (GDP) to measure the significance of the banking sector (Bank) (Demirgüç-Kunt and Maksimovic, 1996).

4. Empirical design

4.1. Target leverage

The existing literature models each firm’s target leverage in a specific country or institutional framework as a function of the firm’s time-varying characteristics, industrial elements, and macroeconomic factors (Frank and Goyal, 2009). Following Flannery and Rangan (2006), Öztekin and Flannery (2012), and An et al. (2015), we regress the observed leverage ratio (LEV) on a set of leverage determinants to estimate the optimal point for both book leverage ratio (BLEV) and market leverage ratio (MLEV). Using this regression, we model the possibility that target leverage might differ across firms or over time

L E V_{i, t + 1, j} = α_{i} + β X_{i, t, j} + f_{i, j} + y_{t, j} + μ_{i, t, j}

(1)

where each firm is indexed by i, time by t, and j by country. $X_{i, t, j}$ is a vector of firm, industry, and macroeconomic variables related to the costs and benefits of operating with various leverage ratios including firm size, tangibility, growth opportunity, profitability, depreciation, research and development, research and development dummy, industry median of leverage, inflation rate, and GDP growth rate. We also control for year fixed effects $y_{t, j}$ and firm fixed effects $f_{i, j}$ in equation (1) to account for unobserved heterogeneity across time and firm.

The target leverage ratio of each firm is measured as the fitted value obtained from regression (1)

L E V_{i, t + 1, j}^{*} = \hat{β} X_{i, t, j} + {\hat{f}}_{i, j} + {\hat{y}}_{t, j}

(2)

4.2. Partial adjustment model of leverage

We next estimate the standard partial adjustment model of capital structure

L E V_{i, t + 1, j} - L E V_{i, t, j} = α_{0} + \partial_{j} (L E V_{i, t + 1, j}^{*} - L E V_{i, t, j}) + ω_{i, t + 1, j}

(3)

where $\partial_{j}$ is a measure of aggregate SOA of firm leverage in a specific country that diverges away from the target of the next period.

While the leverage adjustment speed $\partial_{j}$ in equation (3) is constant for all firms in a specific country, to test our hypotheses, we allow liquidity to affect the firm’s SOA toward its target ratio. Öztekin and Flannery (2012) show that firm accounting variables may affect both target leverage and leverage SOA. We use a set of covariates that are used in target leverage estimation (vector $X_{i, t, j}$ ). Thus, $\partial_{j}$ varies with liquidity and control variables

\partial_{i, t, j} = \partial_{0} + β_{1} L I Q_{i, t, j} + θ X_{i, t, j}

(4)

where $L I Q_{i, t}$ is proxied by liquidity measures including the Amihud measure $(A m i h u d_{i, t})$ , Turnover $(T u r n_{i, t})$ and High-Low impact $(H i g h L o w_{i, t, j})$ .⁵

Substituting equation (4) back to equation (3) yields the equation for a partial adjustment model with heterogeneity in the leverage SOA

Δ L E V_{i, t + 1, j} = α_{0} + (\partial_{0} + β_{1} L I Q_{i, t, j} + θ X_{i, t, j}) (D i s t_{i, t, j}) + ω_{i, t + 1, j}

(5)

where $Δ L E V_{i, t + 1, j} = L E V_{i, t + 1, j} - L E V_{i, t, j}$ and $D i s t_{i, t, j} = L E V_{i, t + 1, j}^{*} - L E V_{i, t, j}$ .

Equation (5) includes a pooled OLS regression of leverage changes on the product of $D i s t_{i, t, j}$ and the main variable affecting leverage SOA (i.e. liquidity) and control variables with bootstrapped standard errors to account for the generated regressors (Çolak et al., 2018; Faulkender et al., 2012; Pagan, 1984).

5. Empirical results

5.1. Descriptive statistics and correlation analysis

Table 1 presents the summary statistics for each country and for the entire sample including descriptive statistics (Panel A) and correlation coefficients of the determinants of target leverage (Panel B). The average firm in the sample has a book leverage ratio of 0.229 and a market leverage of 0.288. In terms of the liquidity measures, the means of Amihud, Turnover, and HighLow are 2.393, 3.757, and 0.031, respectively. Firms in Korea have the highest liquidity while firms in Greece have the lowest liquidity. Panel B reports the correlations between the determinants of the target leverage ratio. In this instance these correlations are low, suggesting that there is little concern with multicollinearity.

Table 1.

Summary statistics.

Panel A: Descriptive statistics
	Country	BLEV	MLEV	PROF	MTB	DEP	SIZE	TANG	R&D	INFL	GGDP	AMIHUD	TURN	HIGHLOW
1	Argentina	0.222	0.327	0.142	1.439	0.048	12.370	0.411	0.000	0.085	0.036	3.370	0.489	0.188
2	Australia	0.199	0.223	0.086	2.165	0.038	12.152	0.313	0.008	0.025	0.029	8.136	2.136	0.025
3	Belgium	0.240	0.308	0.116	2.002	0.052	13.150	0.283	0.017	0.019	0.016	3.755	1.323	0.023
4	Canada	0.195	0.226	0.063	2.051	0.055	12.623	0.476	0.019	0.019	0.021	3.826	2.334	0.357
5	Chile	0.230	0.300	0.110	1.414	0.038	12.874	0.444	0.000	0.034	0.041	0.019	0.811	0.010
6	China	0.259	0.200	0.081	3.828	0.028	12.974	0.326	0.005	0.024	0.091	0.002	14.555	0.036
7	Colombia	0.155	0.262	0.107	1.001	0.031	13.797	0.449	0.000	0.063	0.037	0.009	1.153	0.012
8	Croatia	0.218	0.359	0.064	1.911	0.039	12.145	0.547	0.000	0.019	0.003	2.084	0.905	0.021
9	Denmark	0.242	0.303	0.100	2.255	0.050	12.360	0.313	0.035	0.019	0.013	0.852	1.996	0.024
10	Finland	0.226	0.309	0.116	2.211	0.052	12.681	0.261	0.029	0.016	0.015	3.295	2.013	0.025
11	France	0.226	0.309	0.095	1.961	0.047	12.822	0.193	0.017	0.014	0.015	5.072	1.846	0.025
12	Germany	0.193	0.254	0.099	2.146	0.054	12.594	0.232	0.030	0.014	0.013	7.209	0.402	0.032
13	Greece	0.327	0.448	0.072	1.653	0.035	11.969	0.366	0.003	0.025	0.003	20.067	1.771	0.044
14	Hong Kong	0.187	0.258	0.045	1.585	0.029	12.524	0.262	0.004	0.021	0.034	1.293	2.714	0.036
15	India	0.323	0.463	0.102	1.601	0.032	10.549	0.383	0.001	0.082	0.073	2.672	1.241	0.048
16	Indonesia	0.303	0.391	0.123	2.538	0.041	12.068	0.404	0.000	0.087	0.047	0.044	1.725	0.030
17	Israel	0.313	0.409	0.096	1.808	0.033	12.144	0.257	0.015	0.018	0.037	2.252	1.350	0.026
18	Italy	0.271	0.376	0.082	1.855	0.047	13.294	0.231	0.005	0.018	0.002	1.227	3.217	0.029
19	Japan	0.229	0.342	0.074	1.335	0.033	12.887	0.308	0.012	0.001	0.008	0.014	3.281	0.023
20	Korea	0.270	0.436	0.077	1.011	0.033	12.822	0.370	0.007	0.027	0.039	0.001	11.857	0.038
21	Malaysia	0.226	0.323	0.083	1.220	0.031	11.630	0.382	0.000	0.024	0.048	9.036	1.983	0.026
22	Mexico	0.257	0.357	0.116	1.713	0.040	13.773	0.434	0.000	0.054	0.025	4.338	1.419	0.020
23	Netherlands	0.230	0.266	0.117	2.558	0.049	13.484	0.240	0.012	0.020	0.020	1.434	3.784	0.026
24	Norway	0.317	0.363	0.074	1.834	0.054	12.888	0.375	0.010	0.020	0.017	1.183	3.654	0.032
25	Pakistan	0.278	0.370	0.159	1.838	0.036	11.730	0.452	0.000	0.087	0.040	0.988	3.147	0.028
26	Peru	0.207	0.388	0.168	2.367	0.046	12.356	0.490	0.000	0.030	0.054	3.878	0.935	0.010
27	Philippines	0.230	0.351	0.095	1.596	0.040	12.010	0.416	0.000	0.043	0.050	2.092	1.044	0.025
28	Singapore	0.202	0.298	0.080	1.366	0.034	11.988	0.302	0.002	0.019	0.053	14.386	2.357	0.028
29	Spain	0.268	0.318	0.105	2.542	0.043	13.687	0.328	0.005	0.022	0.021	0.588	2.937	0.027
30	Sweden	0.199	0.229	0.088	2.969	0.045	12.395	0.198	0.026	0.011	0.023	0.644	3.041	0.031
31	Switzerland	0.205	0.242	0.110	2.470	0.045	13.297	0.296	0.029	0.005	0.019	0.726	2.136	0.023
32	Thailand	0.263	0.316	0.122	1.588	0.047	11.566	0.405	0.000	0.024	0.033	2.292	4.700	0.025
33	Turkey	0.230	0.289	0.121	1.883	0.040	12.231	0.344	0.003	0.150	0.051	0.344	12.001	0.036
34	UK	0.180	0.211	0.069	2.527	0.044	12.365	0.277	0.023	0.020	0.020	2.817	3.009	0.031
35	US	0.210	0.194	0.085	3.071	0.048	13.305	0.264	0.052	0.020	0.021	0.740	0.877	0.039
36	All sample	0.229	0.288	0.084	2.101	0.039	12.676	0.309	0.018	0.021	0.029	2.393	3.757	0.031
Panel B: Correlation coefficients
	BLEV	MLEV	SIZE	TANG	MTB	PROF	DEP	R&D	RDDUM	INDMED (BLEV)	INDMED (MLEV)	AMIHUD	TURN	HIGHLOW
BLEV	1
MLEV	0.7888*	1
SIZE	0.1791*	0.1166*	1
TANG	0.2947*	0.2737*	0.0816*	1
MTB	0.0014	–0.2497*	0.0322*	–0.0904*	1
PROF	–0.0819*	–0.1252*	0.1788*	0.0938*	0.0532*	1
DEP	0.0861*	0.0110*	–0.0021	0.3023*	0.0401*	0.1299*	1
RD	–0.1406*	–0.1891*	–0.0720*	–0.1947*	0.1467*	–0.3038*	0.0682*	1
RDDUM	–0.1195*	–0.1172*	0.1634*	–0.1902*	0.0366*	–0.0493*	–0.0016	0.3744*	1
INDMED (BLEV)	0.6496*	0.5663*	0.1479*	0.3212*	–0.0556*	0.0080*	0.0581*	–0.2121*	–0.1842*	1
INDMED (MLEV)	0.5297*	0.7228*	0.0884*	0.2799*	–0.2375*	–0.0479*	–0.0108*	–0.2219*	–0.1448*	0.7848*	1
AMIHUD	0.0418*	0.1239*	–0.2479*	0.0113*	–0.0775*	–0.0987*	0.0363*	–0.0359*	–0.1123*	0.0390*	0.1084*	1
TURN	0.0764*	0.0162*	0.0063*	–0.0024	0.0737*	–0.0547*	–0.0805*	–0.0498*	0.0661*	0.0390*	–0.0045	–0.0985*	1
HIGHLOW	0.0726*	0.0526*	–0.1923*	–0.0688*	0.0707*	–0.2943*	0.0801*	0.1791*	0.0524*	–0.0174*	–0.0362*	0.1156*	0.2446*	1

This table reports the means of firm-level variables by country and for the entire sample. The study period is from 1996 to 2016. The variable definitions are given in Appendix A1 in Online Supplemental Material.

5.2. Liquidity and leverage SOA

We present the baseline regression results (equation (5)), which determine the stock liquidity–SOA relationship (H1), in Table 2. All these regressions are estimated using the pooled OLS method with bootstrapped standard errors. The results are presented separately for book leverage (BLEV) and market leverage (MLEV), as indicated by the column headings. Our independent variable of interest is the interaction term between book (market) target deviation and liquidity, which is proxied by the Amihud (columns 1 and 2), turnover (columns 3 and 4), and high-low impact (columns 5 and 6).

Table 2.

The effect of liquidity on leverage SOA—baseline results.

Variables	Amihud		Turn		HighLow
	ΔBLEV _t+1(1)	ΔMLEV _t+1(2)	ΔBLEV _t+1(3)	ΔMLEV _t+1(4)	ΔBLEV _t+1(5)	ΔMLEV _t+1(6)
Dist × (−Amihud)	0.0006**	0.0006***
	(2.383)	(3.012)
Dist × Turn			0.0016***	0.0034***
			(6.682)	(12.43)
Dist × HighLow					2.1130***	2.276***
					(13.56)	(15.96)
Dist	0.2720***	0.1920***	0.2550***	0.1670***	0.1490***	0.0716***
	(20.71)	(13.65)	(22.32)	(14.24)	(11.86)	(4.607)
Dist × Size	–0.0128***	–0.0069***	–0.0119***	–0.0058***	–0.0086***	–0.0031***
	(–12.67)	(–6.519)	(–14.30)	(–6.283)	(–8.379)	(–2.615)
Dist × MTB	0.0026***	0.0040***	0.0026***	0.0038***	0.0024***	0.0037***
	(4.855)	(3.303)	(5.574)	(3.311)	(4.880)	(3.801)
Dist × Prof	0.2470***	0.0219	0.2570***	0.0429**	0.3190***	0.1170***
	(9.186)	(0.758)	(11.26)	(2.133)	(13.97)	(4.729)
Dist × RD	0.21***	0.1960**	0.2400***	0.2290***	0.1990***	0.1650**
	(3.632)	(2.286)	(5.208)	(2.858)	(3.520)	(2.020)
Dist × RDDum	–0.0075**	–0.0180***	–0.0010**	–0.0232***	–0.0081*	–0.0158***
	(–2.190)	(–4.035)	(–2.505)	(–6.019)	(–1.845)	(–3.621)
Dist × Tang	–0.0568***	–0.0374***	–0.0567***	–0.0364***	–0.0432***	–0.0238**
	(–5.815)	(–3.310)	(–6.013)	(–3.325)	(–4.179)	(–2.377)
Dist × Dep	–0.1400*	0.5250***	–0.1280*	0.5450***	–0.3460***	0.2610***
	(–1.689)	(6.880)	(–1.844)	(7.551)	(–5.007)	(3.031)
Dist × Infl	0.0025***	0.0049***	0.0026***	0.0047***	0.0008	0.0027***
	(3.543)	(7.264)	(3.946)	(6.928)	(1.249)	(3.630)
Dist × GGDP	0.0049***	0.0105***	0.0041***	0.0091***	0.0050***	0.0106***
	(8.812)	(16.28)	(6.012)	(11.76)	(8.771)	(14.15)
Constant	0.0150***	0.0129**	0.0150***	0.0131***	0.0153***	0.0137*
	(3.934)	(2.014)	(3.400)	(2.731)	(3.766)	(1.690)
Observations	190,754	190,754	190,754	190,754	190,754	190,754
R ²	0.057	0.113	0.057	0.114	0.060	0.116
Year FE	Yes	Yes	Yes	Yes	Yes	Yes
Country FE	Yes	Yes	Yes	Yes	Yes	Yes

SOA: speed of adjustment; FE: fixed effect.

This table reports the regression results for the effect of liquidity, proxied by Amihud illiquidity, Turnover, and High-low impact on the leverage speed of adjustment $Δ L E V_{i, t + 1, j} = α_{0} + (\partial_{0} + β_{1} L I Q_{i, t, j} + θ X_{i, t, j}) (D i s t_{i, t, j}) + ω_{i, t + 1, j}$

The coefficients of these interaction terms are positive and highly significant for both book and market leverage regressions, which indicates a positive relation between stock liquidity and leverage SOA. Our findings imply that firms with high liquidity not only have a lower net cost of equity but also lower adverse selection problems, resulting in lower transaction costs, lower overall adjustment costs, and high leverage SOA. The result is consistent with our expectations and with the findings of prior empirical studies of US markets (e.g. Lipson and Mortal, 2009; Zhou et al., 2016).

The coefficients of target distance $(D i s t_{i, t, j})$ are positive and highly significant at the 1% level across models, with magnitudes ranging between 0.072 (column 6) and 0.272 (column 1). This result implies that the base SOA, which does not consider the effects of liquidity, firm size, tangibility, growth opportunity, profitability, depreciation, research and development, research and development dummy, industry median of leverage, inflation rate, and GDP growth rate, is about 7.2%–27.2%. These results are consistent with the findings of dynamic trade-off studies in capital structure literature (Fischer et al., 1989; Goldstein et al., 2001; Strebulaev, 2007).

5.3. Liquidity and leverage SOA for over- and under-levered firms

Next, to analyze the difference in liquidity–leverage SOA sensitivity for over- and under-levered firms, we examine equation (5) separately for each subsample. Panel A of Table 3 reports the results. Specifically, the coefficients of the interaction terms between leverage deviation and liquidity proxies (Dist × (−Amihud), Dist × Turn and Dist × HighLow) across models (1)–(6) are positive and statistically significant at the 1% level for both book and market leverage regressions, indicating that liquidity has positive impacts on the leverage SOA of over-levered firms. This is consistent with our expectations that highly liquid over-levered firms can easily increase their equity, then decrease the leverage ratio and adjust more quickly to their target leverage.

Table 3.

The effect of liquidity on leverage SOA—over- and under-levered firms.

Panel A: The effects of liquidity on leverage SOA for over-levered and under-levered firms
Variables	Over-levered firms						Under-levered firms
	Amihud		Turn		HighLow		Amihud		Turn		HighLow
	ΔBLEV_t+1(1)	ΔMLEV_t+1 (2)	ΔBLEV_t+1 (3)	ΔMLEV_t+1(4)	ΔBLEV_t+1 (5)	ΔMLEV_t+1(6)	ΔBLEV_t+1(7)	ΔMLEV_t+1 (8)	ΔBLEV_t+1(9)	ΔMLEV_t+1 (10)	ΔBLEV_t+1 (11)	ΔMLEV_t+1 (12)
Dist × (−Amihud)	0.0010***	0.0023***					0.0003	0.0018***
	(3.46)	(9.61)					(1.21)	(4.94)
Dist × Turn			0.0023***	0.0049***					0.0007	0.0008
			(6.30)	(10.11)					(1.53)	(1.36)
Dist × HighLow					2.0699***	0.9233***					0.9722***	0.2821
					(10.04)	(3.99)					(4.27)	(1.48)
Dist	0.5173***	0.4976***	0.4848***	0.4110***	0.3330***	0.1338***	0.1052***	0.1148***	0.0479**	0.1042***	0.0940***	0.0514***
	(24.04)	(17.69)	(25.10)	(15.45)	(15.92)	(7.14)	(5.28)	(6.44)	(2.07)	(5.81)	(5.91)	(4.41)
Dist × Size	–0.0316***	–0.0278***	–0.0299***	–0.0231***	0.0222***	0.0110***	–0.0126***	–0.0125***	–0.0098***	–0.0101***	–0.0114***	–0.0056***
	(–20.10)	(–13.19)	(–20.41)	(–12.50)	(13.99)	(8.20)	(–7.29)	(–8.13)	(–5.32)	(–6.08)	(–7.00)	(–7.66)
Dist × MTB	0.0009	0.0103***	0.0010	0.0103***	–0.0000	0.0001	–0.0002	0.0007**	–0.0001	0.0007***	–0.0001	–0.0001
	(1.62)	(3.55)	(1.24)	(3.95)	(–0.26)	(0.50)	(–0.72)	(2.44)	(–0.49)	(2.98)	(–0.53)	(–0.67)
Dist × Prof	0.8728***	0.5531***	0.8847***	0.5856***	–0.6489***	–0.3958***	0.2444***	0.2208***	0.2400***	0.2262***	0.2265***	0.1753***
	(19.04)	(9.86)	(25.87)	(10.73)	(–9.65)	(–2.94)	(4.98)	(7.91)	(4.68)	(5.47)	(3.57)	(5.25)
Dist × RD	0.4929***	0.3750***	0.5249***	0.4121***	0.3782***	0.3066**	0.0332	–0.3371***	–0.0910	–0.3125***	–0.1063	–0.0642
	(7.06)	(3.19)	(6.71)	(2.91)	(3.98)	(2.45)	(0.47)	(–6.01)	(–1.03)	(–6.93)	(–1.20)	(–0.87)
Dist × RDDum	–0.0027	–0.0041	–0.0062	–0.0101	–0.0020	–0.0056	–0.0260***	–0.0105**	–0.0045	–0.0317***	–0.0050	–0.0062**
	(–0.42)	(–0.47)	(–0.91)	(–1.55)	(–0.30)	(–1.39)	(–5.97)	(–2.26)	(–0.88)	(–5.31)	(–0.89)	(–2.37)
Dist × Tang	–0.0864***	–0.0951***	–0.0883***	–0.0954***	–0.0886***	–0.0488***	0.0266**	0.0879***	0.0461***	0.0881***	0.0512***	0.0344***
	(–6.33)	(–6.13)	(–5.80)	(–7.09)	(–5.75)	(–4.84)	(2.37)	(7.84)	(3.43)	(5.51)	(5.06)	(4.69)
Dist × Dep	–0.8865***	0.3840***	–0.8833***	0.3790***	–0.7930***	–0.2764*	0.0640	–0.3316***	–0.1465**	–0.2419***	–0.2039**	–0.0711
	(–9.51)	(3.10)	(–8.81)	(3.43)	(–5.66)	(–1.78)	(0.68)	(–4.64)	(–2.00)	(–2.85)	(–2.32)	(–1.05)
Dist × Infl	0.0026***	0.0021	0.0025**	0.0017*	0.0021*	–0.0005	0.0037***	0.0034***	0.0001	0.0070***	–0.0004	–0.0004
	(2.69)	(1.47)	(2.34)	(1.90)	(1.84)	(–0.57)	(5.03)	(3.45)	(0.16)	(5.95)	(–0.36)	(–0.70)
Dist × GGDP	0.0057***	0.0079***	0.0047***	0.0060***	0.0064***	0.0042***	0.0055***	0.0022***	0.0017**	0.0039***	0.0020**	0.0011**
	(5.93)	(5.43)	(3.94)	(4.07)	(7.75)	(3.73)	(7.07)	(2.75)	(2.01)	(5.12)	(2.53)	(2.03)
Constant	0.0067	–0.0045	0.0059	–0.0070	0.0040	0.0012	0.0188***	0.0322***	0.0267***	0.0231***	0.0273***	0.0279***
	(0.95)	(–0.42)	(1.03)	(–0.54)	(0.55)	(0.19)	(11.86)	(4.57)	(6.75)	(21.99)	(6.16)	(5.77)
Observations	85,624	83,017	85,624	83,017	85,624	85,624	105,130	107,737	105,130	107,737	105,130	105,130
R ²	0.0786	0.1287	0.0790	0.1294	0.0775	0.0324	0.0192	0.0670	0.0270	0.0617	0.0275	0.0205
Year FE	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Country FE	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Panel B: The effects of liquidity on leverage SOA for over-levered firms conditional on the impacts of bankruptcy protection
Variables	Amihud				Turn				HighLow
	ΔBLEV_t+1(1)	ΔMLEV_t+1 (2)	ΔBLEV_t+1 (3)	ΔMLEV_t+1(4)	ΔBLEV_t+1 (5)	ΔMLEV_t+1(6)	ΔBLEV_t+1(7)	ΔMLEV_t+1 (8)	ΔBLEV_t+1(9)	ΔMLEV_t+1 (10)	ΔBLEV_t+1 (11)	ΔMLEV_t+1 (12)
Dist × (−Amihud)	0.0011***		0.0028***
	(5.19)		(10.55)
Dist × (−Amihud) × Bankrp	–0.0008**		–0.0015***
	(–2.08)		(–2.65)
Dist × Turn					0.0011***		0.0080***
			(2.85)		(8.99)
Dist × Turn × Bankrp			–0.0009*		–0.0048***
			(–1.83)		(–4.93)
Dist × HighLow									1.9831***		0.9100***
									(11.10)		(3.98)
Dist × HighLow × Bankrp									–0.3458**		–0.1978
									(–2.13)		(–0.72)
Dist	0.4422***		0.4982***		0.4104***		0.4016***		0.3604***		0.2363***
	(26.48)		(22.39)		(25.64)		(16.00)		(11.27)		(10.39)
Dist × Bankrp	0.0028		–0.0036		–0.0013		0.0342***		–0.0002		0.0635***
	(0.56)		(–0.39)		(–0.24)		(3.56)		(–0.03)		(5.47)
Dist × Size	–0.0271***		–0.0282***		–0.0246***		–0.0246***		–0.0262***		–0.0200***
	(–21.74)		(–18.50)		(–20.18)		(–13.78)		(–13.26)		(–13.07)
Dist × MTB	0.0001		0.0176***		0.0002		0.0173***		0.0008		0.0176***
	(0.23)		(15.84)		(0.63)		(12.75)		(0.93)		(21.89)
Dist × Prof	0.8090***		0.5502***		0.8199***		0.5838***		0.9239***		0.5929***
	(49.51)		(8.39)		(50.26)		(10.38)		(42.31)		(25.30)
Dist × RD	0.4593***		0.3246**		0.4837***		0.3591***		0.4799***		0.3668***
	(14.91)		(2.15)		(15.63)		(2.69)		(6.13)		(7.76)
Dist × RDDum	–0.0063		–0.0057		–0.0086*		–0.0121*		–0.0058		–0.0013
	(–1.38)		(–0.86)		(–1.88)		(–1.66)		(–1.03)		(–0.23)

Panel B: The effects of liquidity on leverage SOA for over-levered firms conditional on the impacts of bankruptcy protection
Variables	Amihud		Turn		HighLow
	ΔBLEV_{t +1} (1)	ΔMLEV_{t +1} (2)	ΔBLEV_{t +1}(3)	ΔMLEV_{t +1} (4)	ΔBLEV_{t +1}(5)	ΔMLEV_{t +1} (6)
Dist × Tang	–0.0660***	–0.0869***	–0.0704***	–0.0877***	–0.0775***	–0.1115***
	(–7.08)	(–6.22)	(–7.57)	(–5.23)	(–3.06)	(–9.46)
Dist × Dep	–0.6473***	0.3546***	–0.6714***	0.3437**	–0.9789***	0.4255***
	(–9.84)	(2.61)	(–10.24)	(2.42)	(–9.45)	(5.11)
Dist × Infl	0.0011	0.0029***	0.0029***	0.0029**	–0.0013	0.0001
	(1.58)	(2.61)	(4.10)	(2.48)	(–1.00)	(0.08)
Dist × GGDP	0.0070***	0.0072***	0.0041***	0.0052***	0.0071***	0.0330***
	(10.03)	(6.46)	(5.21)	(4.30)	(9.04)	(37.67)
Constant	0.0016***	0.0060	0.0148***	0.0050	0.0005	0.0053***
	(4.15)	(1.21)	(7.55)	(1.06)	(0.98)	(7.92)
Observations	81,539	81,539	81,539	81,539	84,168	81,539
R ²	0.0731	0.1328	0.0784	0.1336	0.0729	0.0731
Year FE	Yes	Yes	Yes	Yes	Yes	Yes
Country FE	Yes	Yes	Yes	Yes	Yes	Yes

SOA: speed of adjustment; FE: fixed effect.

This table reports the regression results for the effect of liquidity on the SOA for over- and under-levered firms (Panel A), and the impacts of bankruptcy protection on liquidity-leverage SOA association for over-levered firms (Panel B). The dependent variable is the change in book and market leverage ratio $(Δ L E V_{i, t + 1, j})$ (∆LEVi, t+1, j). $D i s t_{i, t, j}$ Disti, t, j is the difference between the target leverage ratio and the actual leverage ratio. Liquidity $L I Q_{i, t, j}$ LIQi, t, j is proxied by Amihud (Amihud), Turnover (Turn), and High-low impact (HighLow) measures. Bankrpi, j is the bankruptcy code. Control variables include firm size (Size), market-to-book ratio (MTB), profitability (Prof), research and development expenses (RD), research and development dummy (RDDum), Tangibility (Tang), Depreciation expenses (Dep), Inflation rate (INFL), and annual GDP growth rate (GGDP). ***, **, and * indicate significance at the 1%, 5%, and 10% levels, respectively. Standard errors are bootstrapped. T-statistics are reported in parenthesis. The variable definitions are given in Appendix A1 in Online Supplemental Material.

For under-levered firms, the coefficients of the interaction terms between leverage deviation and liquidity measures (Dist × (−Amihud), Dist × Turn, and Dist × HighLow) are statistically insignificant for models (7), (9), (10), and (12) and statistically significant for models (8) and (11). The magnitudes of these coefficients are also smaller than the coefficients of the interaction terms between leverage deviation and liquidity proxies for over-levered firms. These results indicate that the impact of liquidity on leverage SOA is ambiguous and less pronounced for under-levered firms, supporting our hypothesis H2 that firms with high liquidity have a lower cost of equity and prefer equity financing when raising capital. Growing, liquid, under-levered firms have great investment opportunities and require external capital. These firms consider the value enhancement gained by issuing more equity to finance investment opportunities outweighs the benefits of adjusting back to target leverages. This result is in line with the findings of Frieder and Martell (2006) and Lipson and Mortal (2009) that high-liquidity firms prefer and largely rely on equity financing. This result is, however, inconsistent with Cheung et al. (2019), who suggest that the impacts of liquidity on the cost of debt are much greater than the cost of equity. Specifically, high liquidity increases information disclosure, reduces uncertainty about the firm’s future cash flow, reduces the default risk and hence, decreases the cost of debt. Thereby, high liquidity will also speed up the leverage SOA of under-levered firms by enhancing these firms’ debt issuance.

Because of the high level of debt, the threat of bankruptcy is generally more severe for over-levered firms. In adjusting to target leverage, over-levered firms substitute equity for debt and decrease their threat of bankruptcy. Indeed, in some countries that have defined bankruptcy procedures with explicit bankruptcy codes, the rights and claims of lenders are specified and limited; reorganization of the firm is facilitated; and debt payments are deferred (Claessens et al., 2003; Davydenko and Franks, 2008). We conjecture that in countries with explicit bankruptcy codes, the threat of bankruptcy of over-levered firms is less severe, which leads to the impacts of liquidity on leverage SOA for over-levered firms being moderated. To examine the impacts of explicit bankruptcy codes on the liquidity–leverage SOA association for over-levered firms, we follow Gao and Zhu (2015) and Fan et al. (2012) in constructing a dummy variable (“bankruptcy code”) based on the country’s legal system that defines bankruptcy procedures (Djankov et al., 2008). This variable takes a value of one for those countries in which an insolvent firm can undergo a court-supervised reorganization proceeding. Panel B of Table 3 reports the results. The coefficients of the triple interaction terms between liquidity measures, target distance, and dummies for explicit bankruptcy codes (Dist × (−Amihud) × Bankrp, Dist × Turn × Bankrp, and Dist × HighLow × Bankrp) across models (1) to (6) are negative and statistically significant, indicating that in countries with explicit bankruptcy codes, the association between liquidity and leverage SOA is less pronounced for firms with higher level of debts relative to target ratios. The implication is that explicit bankruptcy codes reduce the threat of bankruptcy for over-levered firms, which lowers the incentive of these firms to substitute equities for debts to adjust back to target leverage, thereby moderating the positive impacts of liquidity on leverage SOA of such firms.

5.4. Impacts of institutional environments on the liquidity–leverage SOA relationship

We further examine the impact of institutional environments on the relationship between liquidity and SOA. To examine this issue, we include the interaction terms between stock liquidity and institutional indicators $(I E_{t, j})$ , including rule of law, risk of expropriation, risk of contract repudiation, level of corruption, creditor rights enforcement, and banking sector, in the SOA specification (equation (4))

\partial_{i, t, j} = \partial_{0} + β_{1} L I Q_{i, t, j} + β_{2} I E_{t, j} + β_{3} I E_{t, j} * L I Q_{i, t, j} + θ X_{i, t, j}

(6)

Substituting equation (6) into equation (3) and simplifying yields

\begin{matrix} Δ L E V_{i, t + 1, j} = α_{0} + (\partial_{0} + β_{1} L I Q_{i, t, j} + β_{2} I E_{t, j} + β_{3} I E_{t, j} * L I Q_{i, t, j} \\ + θ X_{i, t, j}) (D i s t_{i, t, j}) + ω_{i, t + 1, j} \end{matrix}

(7)

Table 4 reports the regression results. The coefficients of the interaction term $D i s t_{i, t, j} * L I Q_{i, t, j}$ remain positive and statistically significant. More importantly, the coefficients of the triple interaction term $D i s t_{i, t, j} * L I Q_{i, t, j} * I E_{t, j}$ , which captures the role of institutional environments in reshaping the liquidity–leverage SOA relationship, are negative and statistically significant for book and market leverage regression across models (1)–(36). Specifically, our first institutional environment proxy is RulLaw. Models (1)–(6) in Table 4 show that all six coefficients of the interaction terms between leverage deviation, equity liquidity, and law and order are significantly negative. This set of results is consistent with our hypothesis H3a that the positive impact of equity liquidity on SOA is weaker in countries with strong law and order. In addition, we examine the role of good governments in reshaping the effect of liquidity on SOA. We define a good government using the variables RisExp, RisCon, and Corrup. Results are summarized in models (7)–(24) of Table 4. In general, the three alternative good government measures deliver consistent results with RulLaw. We find that the interactions between leverage deviation, liquidity, and the three good government measures have significantly positive coefficients. There is thus strong evidence that a good government moderates the positive relationship between liquidity and leverage SOA. We next measure institutional environments using CreEnf (models (25)–(30)) and Bank (models (31)–(32)). The results are consistent with previous proxies for institutional environments. Specifically, the interaction terms between leverage deviation, liquidity, and creditor right enforcement and between leverage deviation, liquidity, and the significance of banking sector have negative coefficients. These results imply that the positive association between liquidity and SOA is less pronounced in countries with strong creditor rights enforcement and sophisticated banking systems (H3a).

Table 4.

The effect of liquidity on leverage SOA—impact of institutional environments.

Variables	IE = RulLaw						IE = RisExp
Variables	Amihud		Turn		HighLow		Amihud		Turn		HighLow
	BLEV_t+1(1)	MLEV_t+1(2)	BLEV_t+1(3)	MLEV_t+1(4)	BLEV_t+1(5)	MLEV_t+1(6)	BLEV_t+1(7)	MLEV_t+1(8)	BLEV_t+1(9)	MLEV_t+1(10)	BLEV_t+1(11)	MLEV_t+1(12)
Dist × (−Amihud)	0.0036***	0.0029***					0.0055***	0.0034***
	(3.97)	(2.69)					(3.78)	(2.77)
Dist × (−Amihud) × IE	–0.0004***	–0.0003**					–0.0006***	–0.0003**
	(–3.22)	(–1.99)					(–3.20)	(–2.16)
Dist × Turn			0.0058***	0.0060***					0.0107***	0.0108***
			(4.39)	(6.08)					(3.43)	(4.59)
Dist × Turn × IE			–0.0004**	–0.0002*					–0.0009***	–0.0008***
			(–2.17)	(–1.79)					(–2.59)	(–2.88)
Dist × HighLow					1.2465***	0.8793**					2.0981***	2.1304***
					(3.50)	(2.13)					(3.04)	(2.70)
Dist × HighLow × IE					–0.1044**	–0.1403***					–0.0085	–0.0051
					(–2.53)	(–2.92)					(–0.11)	(–0.06)
Dist	0.2098***	0.1176***	0.1519***	0.0557***	0.1158***	0.0760***	0.2371***	0.1554***	0.1493***	0.0633***	0.1123***	0.0542*
	(12.75)	(5.61)	(8.52)	(3.66)	(6.51)	(3.65)	(10.12)	(6.81)	(5.55)	(2.81)	(3.91)	(1.66)
Dist × IE	0.0086***	0.0132***	0.0130***	0.0173***	0.0041**	0.0057***	0.0035	0.0080***	0.0110***	0.0144***	0.0029	0.0062*
	(6.16)	(7.59)	(7.34)	(14.60)	(2.35)	(2.86)	(1.40)	(3.68)	(4.05)	(6.65)	(0.96)	(1.85)
Dist × Size	–0.0140***	–0.0104***	–0.0133***	–0.0095***	–0.0091***	–0.0070***	–0.0127***	–0.0097***	–0.0119***	–0.0082***	–0.0080***	–0.0057***
	(–13.21)	(–7.71)	(–13.63)	(–9.38)	(–10.19)	(–6.69)	(–11.68)	(–9.14)	(–9.91)	(–7.92)	(–9.00)	(–5.48)
Dist × MTB	0.0032***	0.0051***	0.0032***	0.0051***	0.0031***	0.0048***	0.0034***	0.0052***	0.0034***	0.0052***	0.0032***	0.0050***
	(5.08)	(8.38)	(5.67)	(8.62)	(9.32)	(7.92)	(5.14)	(8.52)	(6.01)	(8.47)	(9.73)	(8.21)
Dist × Prof	0.2278***	0.0306	0.2413***	0.0545***	0.3013***	0.1160***	0.2287***	0.0258*	0.2403***	0.0478***	0.2988***	0.1135***
	(9.43)	(1.29)	(12.09)	(3.99)	(26.96)	(7.97)	(10.60)	(1.83)	(10.20)	(3.39)	(26.73)	(7.80)
Dist × RD	0.1780***	0.1586**	0.2016***	0.1892***	0.1595***	0.1036***	0.2043***	0.1575***	0.2271***	0.1925***	0.1867***	0.1350***
	(3.16)	(2.19)	(3.88)	(5.34)	(6.67)	(2.83)	(4.09)	(4.32)	(3.97)	(5.28)	(7.83)	(3.69)
Dist × RDDum	–0.0110**	–0.0164***	–0.0148***	–0.0219***	–0.0106***	–0.0144***	–0.0120***	–0.0203***	–0.0158***	–0.0261***	–0.0116***	–0.0173***
	(–2.29)	(–3.34)	(–3.15)	(–5.97)	(–3.19)	(–3.82)	(–2.62)	(–5.36)	(–3.13)	(–6.87)	(–3.47)	(–4.57)
Dist × Tang	–0.0515***	–0.0262***	–0.0510***	–0.0232***	–0.0407***	–0.0108	–0.0579***	–0.0288***	–0.0581***	–0.0284***	–0.0455***	–0.0185**
	(–4.60)	(–2.62)	(–6.11)	(–3.02)	(–6.08)	(–1.36)	(–6.30)	(–3.62)	(–5.60)	(–3.57)	(–6.79)	(–2.33)
Dist × Dep	–0.2347***	0.3546***	–0.2382***	0.3474***	–0.4359***	0.1327**	–0.1695**	0.4549***	–0.1666**	0.4542***	–0.3625***	0.2302***
	(–3.26)	(4.22)	(–2.71)	(6.39)	(–9.21)	(2.31)	(–2.27)	(8.13)	(–2.27)	(8.14)	(–7.67)	(4.03)
Dist × Infl	0.0047***	0.0072***	0.0051***	0.0075***	0.0027***	0.0048***	0.0031***	0.0054***	0.0033***	0.0054***	0.0012**	0.0031***
	(7.56)	(7.47)	(7.97)	(12.96)	(5.00)	(7.91)	(3.79)	(8.96)	(4.35)	(9.04)	(2.27)	(5.04)
Dist × GGDP	0.0077***	0.0132***	0.0079***	0.0130***	0.0082***	0.0124***	0.0064***	0.0107***	0.0066***	0.0105***	0.0070***	0.0111***
	(12.00)	(19.16)	(10.88)	(21.25)	(14.55)	(19.60)	(8.85)	(16.79)	(7.54)	(16.67)	(12.44)	(17.61)
Constant	0.0153***	0.0141**	0.0153***	0.0121***	0.0155***	0.0018***	0.0152***	0.0039	0.0154***	0.0046	0.0141***	0.0019***
	(3.46)	(2.22)	(3.41)	(7.44)	(3.82)	(6.58)	(4.64)	(0.70)	(3.85)	(0.82)	(12.15)	(6.65)
Observations	176,108	176,108	176,108	176,108	176,108	176,108	176,108	176,108	176,108	176,108	176,108	176,108
R ²	0.0577	0.1126	0.0584	0.1116	0.0610	0.0536	0.0572	0.0531	0.0578	0.0542	0.0586	0.0532
Year FE	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Country FE	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Variables	IE = RisCon						IE = Corrup
	Amihud		Turn		HighLow		Amihud		Turn		HighLow
	BLEV_t+1(13)	MLEV_t+1(14)	BLEV_t+1(15)	MLEV_t+1(16)	BLEV_tt+1(17)	MLEV_t+1(18)	BLEVt+1(19)	MLEV_t+1(20)	BLEV_t+1(21)	MLEV_t+1(22)	BLEV_t+1(23)	MLEV_t+1(24)
Dist × (−Amihud)	0.0052***	0.0020**					0.0040***	0.0040***
	(3.48)	(2.03)					(3.39)	(3.17)
Dist × (−Amihud) × IE	–0.0006***	–0.0002					–0.0004***	–0.0004**
	(–2.99)	(–1.39)					(–2.73)	(–2.44)
Dist × Turn			0.0058***	0.0095***					0.0055***	0.0050***
			(4.13)	(4.20)					(3.64)	(4.72)
Dist × Turn × IE			–0.0004**	–0.0007***					–0.0004*	–0.0002
			(–2.29)	(–2.75)					(–1.80)	(–0.96)
Dist × HighLow					2.8554***	1.7195**					0.8408**	0.4261
					(3.97)	(2.57)					(2.23)	(0.66)
Dist × HighLow × IE					–0.5782***	–0.4446***					–0.1643***	–0.2134***
					(–6.77)	(–5.86)					(–3.54)	(–2.68)
Dist	0.2177***	0.1420***	0.1652***	0.0676***	0.2028***	0.0894***	0.2000***	0.1175***	0.1424***	0.0543***	0.1141***	0.0535**
	(8.08)	(2.87)	(5.05)	(3.83)	(6.09)	(3.10)	(11.41)	(5.33)	(6.76)	(2.88)	(6.17)	(2.17)
Dist × IE	0.0056**	0.0094**	0.0090***	0.0125***	–0.0075**	0.0000	0.0092***	0.0125***	0.0130***	0.0159***	0.0031*	0.0050*
	(2.43)	(2.08)	(2.65)	(7.66)	(–2.13)	(0.02)	(6.42)	(7.84)	(7.92)	(12.44)	(1.68)	(1.77)
Dist × Size	–0.0127***	–0.0097***	–0.0117***	–0.0072***	–0.0084***	–0.0051***	–0.0134***	–0.0096***	–0.0122***	–0.0080***	–0.0085***	–0.0052***
	(–11.77)	(–6.48)	(–9.83)	(–6.61)	(–8.73)	(–5.02)	(–11.56)	(–7.90)	(–10.12)	(–7.77)	(–9.62)	(–5.28)
Dist × MTB	0.0034***	0.0053***	0.0035***	0.0055***	0.0031***	0.0050***	0.0033***	0.0053***	0.0034***	0.0053***	0.0031***	0.0050***
	(5.88)	(7.80)	(6.03)	(5.80)	(5.11)	(8.51)	(6.04)	(6.62)	(6.83)	(7.16)	(9.50)	(5.54)
Dist × Prof	0.2306***	0.0278	0.2425***	0.0475***	0.3071***	0.1221***	0.2310***	0.0330	0.2449***	0.0562*	0.3057***	0.1236***
	(8.91)	(1.52)	(15.93)	(3.40)	(12.07)	(8.66)	(11.32)	(1.37)	(11.33)	(1.77)	(27.35)	(6.13)
Dist × RD	0.2112***	0.1689**	0.2415***	0.2314***	0.1847***	0.1643***	0.1912***	0.1719**	0.2213***	0.2066**	0.1711***	0.1404**
	(4.46)	(2.02)	(6.35)	(3.18)	(3.47)	(4.66)	(3.74)	(2.00)	(4.54)	(2.47)	(7.23)	(2.05)
Dist × RDDum	–0.0126***	–0.0215***	–0.0163***	–0.0254***	–0.0131***	–0.0182***	–0.0108**	–0.0162***	–0.0144***	–0.0212***	–0.0101***	–0.0128***
	(–3.06)	(–3.40)	(–5.87)	(–4.97)	(–2.93)	(–4.96)	(–2.52)	(–3.11)	(–3.19)	(–5.08)	(–3.05)	(–2.61)
Dist × Tang	–0.0570***	–0.0289*	–0.0591***	–0.0362***	–0.0404***	–0.0208***	–0.0523***	–0.0292***	–0.0526***	–0.0287***	–0.0408***	–0.0197**
	(–5.00)	(–1.79)	(–5.54)	(–3.39)	(–3.89)	(–2.70)	(–5.75)	(–3.22)	(–4.66)	(–3.55)	(–6.11)	(–2.08)
Dist × Dep	–0.1680**	0.4760***	–0.1429**	0.4909***	–0.4184***	0.1883***	–0.2009***	0.4114***	–0.1926**	0.4180***	–0.4187***	0.1545**
	(–1.99)	(4.27)	(–2.09)	(9.02)	(–5.41)	(3.46)	(–2.97)	(5.73)	(–2.47)	(4.59)	(–8.94)	(2.33)
Dist × Infl	0.0037***	0.0060***	0.0038***	0.0057***	0.0032***	0.0049***	0.0049***	0.0071***	0.0052***	0.0073***	0.0030***	0.0050***
	(4.76)	(4.28)	(3.95)	(7.50)	(3.60)	(7.77)	(7.92)	(9.59)	(7.99)	(9.55)	(5.48)	(6.10)
Dist × GGDP	0.0068***	0.0110***	0.0069***	0.0116***	0.0080***	0.0130***	0.0075***	0.0126***	0.0075***	0.0124***	0.0081***	0.0132***
	(8.09)	(11.91)	(15.34)	(13.03)	(10.08)	(21.01)	(11.52)	(16.41)	(10.98)	(19.06)	(14.56)	(16.93)
Constant	0.0153***	0.0041	0.0154***	0.0147**	0.0155***	0.0149***	0.0151***	0.0138**	0.0152***	0.0141**	0.0154***	0.0145**
	(3.85)	(0.77)	(3.30)	(2.03)	(3.26)	(2.63)	(4.28)	(2.21)	(4.16)	(2.33)	(3.80)	(2.04)
Observations	176,108	176,108	176,108	176,108	176,108	176,108	176,108	176,108	176,108	176,108	176,108	176,108
R ²	0.0573	0.0531	0.0578	0.1129	0.0612	0.1148	0.0576	0.1125	0.0583	0.1135	0.0610	0.1150
Year FE	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Country FE	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Variables	IE = CreEnf						IE = Bank
	Amihud		Turn		HighLow		Amihud		Turn		HighLow
	BLEV_t+1(25)	MLEV_t+1(26)	BLEV_t+1(27)	MLEV_t+1(28)	BLEV_t+1(29)	MLEV_t+1(30)	BLEV_t+1(31)	MLEV_t+1(32)	BLEV_t+1(33)	MLEVt+1(34)	BLEV_t+1(35)	MLEV_t+1(36)
Dist × (−Amihud)	0.0015***	0.0032***					0.0013***	0.0014***
	(2.69)	(3.47)					(6.74)	(6.98)
Dist × (−Amihud) × IE	–0.0003	–0.0009***					–0.0019***	–0.0021***
	(–1.59)	(–2.85)					(–3.41)	(–3.77)
Dist × Turn			0.0063***	0.0054***					0.0016***	0.0033***
			(3.30)	(2.70)					(3.92)	(9.78)
Dist × Turn × IE			–0.0014**	–0.0006					–0.0002	–0.0002
			(–2.29)	(–0.92)					(–0.56)	(–0.36)
Dist × HighLow					5.4296***	4.5549***					1.9680***	1.8915***
					(8.85)	(12.84)					(12.01)	(9.76)
Dist × HighLow × IE					–1.2212***	–0.8385***					–0.1782	–0.5603***
					(–6.01)	(–6.94)					(–1.08)	(–3.17)
Dist	0.3676***	0.2664***	0.3384***	0.2339***	0.0992***	0.0402**	0.3032***	0.2176***	0.2775***	0.1837***	0.1735***	0.0969***
	(28.40)	(14.54)	(17.15)	(12.84)	(4.59)	(2.08)	(18.34)	(13.81)	(15.97)	(15.60)	(8.98)	(5.94)
Dist × IE	–0.0345***	–0.0251***	–0.0320***	–0.0232***	0.0160**	0.0151***	–0.0435***	–0.0314***	–0.0360***	–0.0231***	–0.0376***	–0.0330***
	(–15.37)	(–9.41)	(–9.04)	(–7.50)	(2.50)	(3.23)	(–9.25)	(–8.02)	(–7.82)	(–6.73)	(–6.49)	(–5.09)
Dist × Size	–0.0131***	–0.0074***	–0.0120***	–0.0062***	–0.0084***	–0.0038***	–0.0120***	–0.0067***	–0.0111***	–0.0055***	–0.0079***	–0.0031**
	(–15.14)	(–7.27)	(–9.74)	(–4.25)	(–8.89)	(–3.78)	(–11.22)	(–6.81)	(–9.84)	(–5.81)	(–6.22)	(–2.51)
Dist × MTB	0.0027***	0.0041***	0.0026***	0.0039***	0.0025***	0.0038***	0.0028***	0.0060***	0.0028***	0.0058***	0.0026***	0.0055***
	(8.82)	(4.34)	(3.90)	(4.79)	(5.30)	(7.60)	(5.37)	(7.53)	(7.20)	(7.81)	(4.50)	(8.66)
Dist × Prof	0.2559***	0.0318	0.2695***	0.0566***	0.3311***	0.1255***	0.2474***	0.0214	0.2583***	0.0435*	0.3182***	0.1151***
	(23.79)	(1.26)	(8.60)	(3.11)	(14.90)	(8.75)	(11.59)	(0.77)	(12.76)	(1.83)	(11.89)	(5.13)
Dist × RD	0.1996***	0.1868**	0.2308***	0.2239***	0.1895***	0.1367***	0.2121***	0.1667**	0.2386***	0.2089***	0.1960***	0.1391**
	(8.52)	(2.33)	(4.74)	(4.62)	(4.10)	(3.82)	(4.41)	(2.16)	(4.54)	(2.71)	(3.29)	(2.00)
Dist × RDDum	–0.0031	–0.0139***	–0.0058	–0.0194***	–0.0039	–0.0144***	–0.0016	–0.0126***	–0.0041	–0.0185***	–0.0029	–0.0116**
	(–0.96)	(–2.88)	(–1.34)	(–4.55)	(–0.90)	(–3.91)	(–0.37)	(–3.23)	(–0.80)	(–3.87)	(–1.01)	(–2.36)
Dist × Tang	–0.0499***	–0.0330***	–0.0497***	–0.0319***	–0.0394***	–0.0171**	–0.0585***	–0.0366***	–0.0591***	–0.0365***	–0.0454***	–0.0232***
	(–7.73)	(–3.33)	(–6.32)	(–2.81)	(–4.82)	(–2.23)	(–5.15)	(–4.12)	(–5.47)	(–3.78)	(–4.83)	(–2.73)
Dist × Dep	–0.1681***	0.5039***	–0.1522**	0.5206***	–0.3742***	0.2297***	–0.1501**	0.4851***	–0.1281*	0.5262***	–0.3445***	0.2324***
	(–3.71)	(7.04)	(–2.36)	(9.16)	(–5.91)	(4.18)	(–2.31)	(5.46)	(–1.84)	(7.77)	(–5.34)	(3.33)
Dist × Infl	0.0034***	0.0054***	0.0033***	0.0053***	0.0017***	0.0030***	–0.0007	0.0026***	–0.0003	0.0029***	–0.0015*	0.0018**
	(7.17)	(6.43)	(5.78)	(7.69)	(2.93)	(5.49)	(–0.85)	(3.26)	(–0.38)	(3.30)	(–1.76)	(2.11)
Dist × GGDP	0.0055***	0.0107***	0.0046***	0.0090***	0.0055***	0.0106***	0.0024***	0.0083***	0.0021***	0.0075***	0.0032***	0.0095***
	(12.43)	(17.11)	(7.82)	(15.40)	(8.74)	(19.91)	(3.79)	(11.13)	(3.01)	(11.14)	(7.56)	(14.80)
Constant	0.0158***	0.0141**	0.0157***	0.0144***	0.0155***	0.0044	0.0146***	0.0118*	0.0147***	0.0122*	0.0150***	0.0127*
	(3.86)	(2.25)	(3.71)	(2.94)	(3.60)	(0.80)	(3.36)	(1.91)	(3.35)	(1.75)	(5.77)	(1.94)
Observations	190,754	190,754	190,754	190,754	190,754	190,754	187,997	187,997	187,997	187,997	187,997	187,997
R ²	0.0583	0.1136	0.0588	0.1145	0.0617	0.0574	0.0583	0.1144	0.0583	0.1150	0.0612	0.1169
Year FE	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Country FE	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes

SOA: speed of adjustment; FE: fixed effect.

This table reports the regression results for the effect information environments on liquidity—leverage SOA relationship using the following models: $Δ L E V_{i, t + 1, j} = α_{0} + (\partial_{0} + β_{1} L I Q_{i, t, j} + β_{2} I E_{t, j} + β_{3} I E_{t, j} * L I Q_{i, t, j} + θ X_{i, t, j}) (D i s t_{i, t, j}) + ω_{i, t + 1, j}$

The dependent variable is the change in book and market leverage ratio $(Δ L E V_{i, t + 1, j})$ (∆LEVi, t+1, j). Disti, t, j is the different between the target leverage ratio and the actual leverage ratio. Liquidity $L I Q_{i, t, j}$ LIQi, t, j is proxied by Amihud (Amihud), Turnover (Turn), and High-low impact (HighLow) measure. IEt, j is the institutional environment indicators including rule of law (RulLawt, j), Risk of Expropriation (RisExpt, j), Risk of Contract Repudiation(RisCont, j), Creditor Right Enforcement(CreEnft, j), Level of Corruption (Corrupt, j), and Banking sector (Bankt, j). Xi, t, j is the vector of control variables that includes firm size (Size), market-to-book ratio (MTB), profitability (Prof), research and development expenses (RD), research and development dummy (RDDum), Tangibility (Tang), Depreciation expenses (Dep), Inflation rate (INFL), and annual GDP growth rate (GGDP). Year and country fixed effects are included in Models (1) to (36). ***, **, and * indicate significance at the 1%, 5%, and 10% levels, respectively. Standard errors are bootstrapped. T-statistics are reported in parenthesis. The variable definitions are given in Appendix A1 in Online Supplemental Material.

In general, the results are consistent with our hypothesis H3a that the positive impact of liquidity on SOA is weaker in countries with high-quality institutional settings. Furthermore, the coefficients of interaction terms $D i s t_{i, t, j} * I E_{t, j}$ are positive and significant across the panels, indicating that firms operating in better institutional environments adjust more rapidly to their target levels. These results further confirm the findings of Öztekin and Flannery (2012), Öztekin (2015), and Çolak et al. (2018).

5.5. Endogeneity problems

Section 5 presents the relationship between liquidity and firms’ leverage SOA. However, these findings may be plagued by potential endogeneity problems, including reverse causality and omitted variables bias. We address these concerns by using a shock to equity liquidity and controlling for additional determining factors of leverage SOA that may be associated with equity liquidity.

5.5.1. The introduction of the Directive on Markets in Financial Instruments

It may be that investors prefer to trade stocks of firms operating closer to their target level and with higher leverage SOAs, as these firms are at their highest value (Fischer et al., 1989). That is, higher leverage SOA causes higher stock liquidity (i.e. reverse causality). We address this potential issue by using an exogenous shock to liquidity.

We follow Cumming et al. (2011) and use the introduction of the Directive on Markets in Financial Instruments (MiFID) in European countries as a plausible exogenous shock to liquidity to assess the causal impact of liquidity on leverage SOA. Specifically, in November 2007, MiFID, a major legislative change in the European Union’s Financial Services Action Plan (FSAP), became effective. This gave rise to more detailed rules and more transparent investor protection for the European exchanges. Cumming et al. (2011) show that firms experienced a substantial increase in stock liquidity after the MiFID. The change in this legislation is likely to be exogenous to firms’ strategies and decisions, and thus unlikely to directly drive leverage SOA. Moreover, because the introduction of MiFID affected only countries in the European Union, it created a natural experimental setting to assess the effect of liquidity on leverage SOA.

To test the impact of a change in liquidity on leverage SOA, we use the difference-in-differences regression using a control group to subtract other changes occurring at the same time as MiFID, assuming these other changes were identical between the treatment group (the European countries that were subject to MiFID) and the control group (the other countries) as in equation (8)

Δ L E V_{i, t + 1, j} = α_{0} + (\partial_{0} + β_{1} T r e a t_{i} * A f t e r_{t} + θ X_{i, t, j}) (D i s t_{i, t, j}) + ω_{i, t + 1, j}

(8)

where $T r e a t$ is a dummy variable that is set as one for countries subject to MiFID and zero otherwise, and $A f t e r$ is a dummy variable that is set as one for every year after 2007 and zero in all prior years.

Table 5 presents the regression results. The coefficients of the interaction terms Dist*Treat*After are positive and statistically significant for both book and market leverage regressions, which is in line with our conjecture that an exogenous increase in liquidity resulting from the introduction of MiFID is associated with a higher leverage SOA. Overall, the results of the natural experimental setting provide some evidence of a causal impact of liquidity on leverage SOA.

Table 5.

The introduction of MiFID as an exogenous shock.

Variables	ΔBLEV_t+1	ΔMLEV_t+1
	(1)	(2)
Dist × Treat × After	0.0061**	0.0264***
	(2.17)	(3.25)
Dist	0.2587***	0.1714***
	(14.12)	(12.57)
Dist × Size	–0.0119***	–0.0058***
	(–10.75)	(–5.22)
Dist × MTB	0.0027***	0.0040***
	(12.82)	(3.34)
Dist × Prof	0.2495***	0.0271
	(1,066.13)	(0.98)
Dist × RD	0.2200	0.1948**
	(1.51)	(2.55)
Dist × RDDum	–0.0067*	–0.0157***
	(–1.82)	(–3.26)
Dist × Tang	–0.0568***	–0.0345***
	(–16.20)	(–4.19)
Dist × Dep	–0.1487***	0.5022***
	(–3.36)	(6.93)
Dist × Infl	0.0025***	0.0047***
	(2.93)	(6.57)
Dist × GGDP	0.0052***	0.0110***
	(7.69)	(15.38)
Constant	0.0150***	0.0130**
	(3.91)	(2.25)
Observations	190,754	190,754
R ²	0.0570	0.1130
Year FE	Yes	Yes
Country FE	Yes	Yes

MiFID: Markets in Financial Instruments; FE: fixed effect.

This table reports the regression results of the following model: $Δ L E V_{i, t + 1, j} = α_{0} + (\partial_{0} + β_{1} T r e a t_{i} * A f t e r_{t} + θ X_{i, t, j}) (D i s t_{i, t, j}) + ω_{i, t + 1, j}$

The dependent variable is the change in book and market leverage ratio $(Δ L E V_{i, t + 1, j})$ (∆LEVi,t+1,j). $D i s t_{i, t, j}$ Disti,t,j is the different between the target leverage ratio and the actual leverage ratio. Treat is a dummy variable that is equal to one for countries subject to MiFID and zero otherwise, and Aftert is a dummy variable that is set to one for every year after 2007 and zero in all prior years. Xi, t, j is the vector of control variables that includes firm size (Size), market-to-book ratio (MTB), profitability (Prof), research and development expenses (RD), research and development dummy (RDDum), Tangibility (Tang), Depreciation expenses (Dep), Inflation rate (INFL), and annual GDP growth rate (GGDP). Year and country fixed effects are included in Models (1) and (2). ***, **, and * indicate significance at the 1%, 5%, and 10% levels, respectively. Standard errors are bootstrapped. T-statistics are reported in parenthesis. The variable definitions are given in Appendix A1 in Online Supplemental Material.

5.5.2. Additional controls

There is another potential source of endogeneity issue pertaining to this study, which is the presence of time-varying omitted variables. Specifically, it is possible that stock liquidity is significantly associated with other firm-level and country-level variables, which are typically correlated with leverage SOA, thus biasing the article’s inference. We address this issue by controlling for additional determinants of leverage SOA that may have significant impacts on liquidity.

We first include stock capitalization and term spread in our leverage SOA regression (equation (4)). Using term spread, defined as the difference between the 10-year government bond yield series and the 3-month interest rate series, as a predictor for good economy, Cook and Tang (2010) propose that firms operating in good economies adjust their leverage faster toward the targets relative to firms operating in bad economies. Öztekin and Flannery (2012) and Çolak et al. (2018) also suggest that if a country has a good sized stock market, firms in that country are likely to have higher SOA. We re-estimate equation (5) using these additional controls and present the results in panel A of Table 6. The coefficients of the interaction terms between liquidity proxies and leverage deviation are all positive and statistically significant for both book and market leverage regressions across models (1)–(6). The effects of macro-level control variables are also positive and statistically significant, implying that firms in countries with better economies adjust faster to their target leverage. These results are consistent with not only our baseline results but also the previous literature.

Table 6.

Different sets of control variables.

Panel A: Controlling for additional macro-level variables
Variables	Amihud		Turn		HighLow
	ΔBLEV_t+1(1)	ΔMLEV_t+1(2)	ΔBLEV_t+1(3)	ΔMLEV_t+1(4)	ΔBLEV_t+1(5)	ΔMLEV_t+1(6)
Dist × (−Amihud)	0.0007**	0.0007***
	(1.97)	(3.31)
Dist × Turn			0.0024***	0.0030***
			(6.01)	(8.01)
Dist × HighLow					1.8174***	1.4731***
					(12.09)	(8.48)
Dist	0.2166***	0.1313***	0.1869***	0.0972***	0.1131***	0.0549***
	(12.71)	(8.48)	(9.41)	(5.40)	(6.32)	(3.26)
Dist × Size	–0.0139***	–0.0105***	–0.0129***	–0.0092***	–0.0086***	–0.0067***
	(–10.84)	(–9.07)	(–9.66)	(–6.72)	(–6.86)	(–6.14)
Dist × MTB	0.0037***	0.0046***	0.0037***	0.0044***	0.0037***	0.0046***
	(5.19)	(6.63)	(5.95)	(4.97)	(6.48)	(4.81)
Dist × Prof	0.2059***	0.0586***	0.2160***	0.0762***	0.2639***	0.1153***
	(7.21)	(3.71)	(7.80)	(3.19)	(8.91)	(3.61)
Dist × RD	0.1402**	0.1226***	0.1638**	0.1498*	0.1463***	0.1195
	(2.52)	(3.49)	(2.34)	(1.82)	(2.66)	(1.63)
Dist × RDDum	0.0002	–0.0053	–0.0025	–0.0094*	–0.0010	–0.0051
	(0.05)	(–1.28)	(–0.46)	(–1.70)	(–0.20)	(–1.09)
Dist × Tang	–0.0236**	–0.0113	–0.0245**	–0.0131	–0.0154	–0.0050
	(–2.01)	(–1.27)	(–2.18)	(–0.88)	(–1.33)	(–0.41)
Dist × Dep	–0.2369**	0.3016***	–0.2260**	0.3171***	–0.4175***	0.1365*
	(–2.29)	(5.12)	(–2.38)	(4.50)	(–5.12)	(1.66)
Dist × Infl	0.0099***	0.0100***	0.0099***	0.0096***	0.0053***	0.0057***
	(6.70)	(8.40)	(6.24)	(5.71)	(2.79)	(4.18)
Dist × StockCap	0.0005***	0.0009***	0.0006***	0.0010***	0.0003***	0.0007***
	(6.62)	(14.24)	(7.28)	(11.97)	(4.62)	(8.27)
Dist × TermSpread	0.0042**	0.0201***	0.0038**	0.0198***	0.0008	0.0167***
	(2.57)	(13.90)	(2.48)	(11.34)	(0.62)	(8.94)
Dist × GGDP	0.0058***	0.0105***	0.0053***	0.0098***	0.0071***	0.0116***
	(5.51)	(11.92)	(6.04)	(8.01)	(8.17)	(8.71)
Constant	0.0027	–0.0103***	0.0028	–0.0101***	0.0028	–0.0103***
	(1.15)	(–3.35)	(1.17)	(–3.46)	(1.22)	(–3.30)
Observations	122,442	122,442	122,442	122,442	122,442	122,442
R ²	0.0624	0.1185	0.0629	0.1192	0.0649	0.1197
Year FE	Yes	Yes	Yes	Yes	Yes	Yes
Country FE	Yes	Yes	Yes	Yes	Yes	Yes
Panel B: Controlling for additional firm-level variables
Variables	Amihud		Turn		HighLow
	ΔBLEV_t+1(1)	ΔMLEV_t+1(2)	ΔBLEV_t+1(3)	ΔMLEV_t+1(4)	ΔBLEVt+1(5)	ΔMLEV_t+1(6)
Dist × (−Amihud)	0.0009***	0.0014***
	(3.27)	(4.47)
Dist × Turn			0.0020***	0.0030***
			(4.26)	(5.68)
Dist × HighLow					1.7989***	1.9630***
					(8.36)	(11.67)
Dist	0.3587***	0.3679***	0.3241***	0.3106***	0.2199***	0.2114***
	(19.29)	(15.53)	(16.98)	(13.01)	(14.82)	(12.47)
Dist × Size	–0.0170***	–0.0141***	–0.0153***	–0.0112***	–0.0118***	–0.0083***
	(–12.20)	(–9.12)	(–12.06)	(–6.43)	(–12.42)	(–7.25)
Dist × MTB	0.0028***	0.0054***	0.0029***	0.0053***	0.0028***	0.0052***
	(4.13)	(5.23)	(4.09)	(5.30)	(4.17)	(6.45)
Dist × Prof	0.2195***	–0.0295	0.2255***	–0.0157	0.2700***	0.0430
	(7.40)	(–0.94)	(6.79)	(–0.43)	(9.54)	(1.32)
Dist × RD	0.2510***	0.2013***	0.2768***	0.2378***	0.2415***	0.1863**
	(4.31)	(2.79)	(4.09)	(2.95)	(4.23)	(2.32)
Dist × RDDum	–0.0147***	–0.0204***	–0.0159***	–0.0221***	–0.0130***	–0.0153***
	(–3.01)	(–3.91)	(–3.04)	(–4.02)	(–2.73)	(–3.04)
Dist × Tang	–0.0604***	–0.0446***	–0.0626***	–0.0465***	–0.0561***	–0.0397***
	(–5.67)	(–3.93)	(–5.66)	(–4.33)	(–4.30)	(–3.64)
Dist × Dep	–0.3231***	0.1997**	–0.3115***	0.2127**	–0.4990***	–0.0190
	(–4.07)	(2.09)	(–3.71)	(2.13)	(–4.50)	(–0.20)
Dist × Surplus	0.0256***	0.0523***	0.0282***	0.0555***	0.0329***	0.0580***
	(4.53)	(7.98)	(5.19)	(8.30)	(5.31)	(8.31)
Dist × UnderLev	–0.0473***	–0.1327***	–0.0442***	–0.1276***	–0.0297***	–0.1099***
	(–7.16)	(–17.58)	(–6.21)	(–21.06)	(–3.85)	(–15.99)
Dist × Infl	0.0018*	0.0024**	0.0021*	0.0030***	0.0004	0.0007
	(1.90)	(2.50)	(1.94)	(2.66)	(0.38)	(0.65)
Dist × GGDP	0.0027***	0.0055***	0.0029***	0.0057***	0.0042***	0.0073***
	(2.67)	(5.88)	(3.42)	(5.54)	(4.59)	(7.94)
Constant	0.0130**	0.0553***	0.0129**	0.0552***	0.0131*	0.0552***
	(2.17)	(5.81)	(2.18)	(4.79)	(1.77)	(6.00)
Observations	110,854	110,854	110,854	110,854	110,854	110,854
R ²	0.0628	0.1213	0.0629	0.1215	0.0647	0.1229
Year FE	Yes	Yes	Yes	Yes	Yes	Yes
Country FE	Yes	Yes	Yes	Yes	Yes	Yes

SOA: speed of adjustment; FE: fixed effect.

This table reports the regression results for the effects of liquidity on the SOA after controlling for additional macro-level variables (Panel A) and firm-level variables (Panel B) for following regression: $Δ L E V_{i, t + 1, j} = α_{0} + (\partial_{0} + β_{1} L I Q_{i, t, j} + θ X_{i, t, j} + β_{2} S t o c k M a r k e t_{i, t, j} + β_{3} B o n d M a r k e t_{i, t, j}) (D i s t_{i, t, j}) + ω_{i, t + 1, j}$ $Δ L E V_{i, t + 1, j} = α_{0} + (\partial_{0} + β_{1} L I Q_{i, t, j} + θ X_{i, t, j} + β_{2} S u r p l u s_{i, t, j} + β_{3} U n d e r L e v_{i, t, j}) (D i s t_{i, t, j}) + ω_{i, t + 1, j}$

The dependent variable is the change in book and market leverage ratio (∆LEVi,t+1,j). Disti,t,j is the different between the target leverage ratio and the actual leverage ratio. Liquidity LIQi, t, j is proxied by Amihud (Amihud), Turnover (Turn), and High-low impact (HighLow) measure. Control variables include vector $X_{i, t, j}$ Xi, t, j that are firm size (Size), market-to-book ratio (MTB), profitability (Prof), research and development expenses (RD), research and development dummy (RDDum), Tangibility (Tang), Depreciation expenses (Dep), Inflation rate (INFL), and annual GDP growth rate (GGDP), and additional variables. The first set of additional control variables including stock market capitalization (StockCap) and term spread (TermSpread) (Panel A). The second set of additional control variables including a financing surplus dummy variable (Surplus), that equals one if the firm has a financing surplus, and zero otherwise and an under-target leverage dummy variable (UnderLev) that equals one if the firm is under-levered relative to target leverage, and zero otherwise (Panel B). Year and country fixed effects are included in Models (1) to (6) in both Panel A and Panel B. ***, **, and * indicate significance at the 1%, 5%, and 10% levels, respectively. Standard errors are bootstrapped. T-statistics are reported in parenthesis. The variable definitions are given in Appendix A1 in Online Supplemental Material.

We next control for additional firm-level factors that are known to have effects on leverage SOA and liquidity. Specifically, Faulkender et al. (2012) argue that deviation from target leverage and cash flow status are two important determinants of leverage SOA. They note that the adjustment costs and thus leverage SOA depend on whether firms are over- or under-levered in relation to target leverage, as well as whether they are deficit or surplus financing. We therefore include these two firm-level control variables in the leverage SOA specification (equation (4)). Panel B of Table 6 presents the results of this additional examination. Consistent with Faulkender et al. (2012), our results show that while the coefficients of the interaction term Dist × Surplus are positive and significant at the 1% level, the coefficients of the interaction term Dist × UnderLev are negative and significant at the 1% level. The results imply that firms with financing surplus adjust faster to their targets and under-levered firms have lower leverage SOA. More importantly, the coefficients of interaction terms between leverage distance and liquidity proxies remain positive and statistically significant, indicating that our baseline results are robust in the presence of additional controlling variables.

6. Conclusion

In this study, we investigate how firm-level liquidity affects the speed with which firms converge back to their leverage targets in an international setting. Based on a sample of 16,963 firms in 35 countries over the period between 1996 and 2016, we find a positive association between liquidity and leverage SOA, indicating that firms with highly liquid equity adjust more quickly to their targets. This important finding is robust after using a range of alternative proxies for liquidity, an alternative leverage measure, an alternative econometric method, and correcting for endogeneity. More interestingly, we show that liquidity has a distinct effect on the leverage SOA of over- and under-levered firms: highly liquid firms that are over-levered tend to adjust quickly to their targets and this positive impact is attenuated in countries with explicit bankruptcy codes, whereas the positive association between liquidity and leverage SOA is ambiguous for under-levered firms. In addition, we confirm a substitution effect between country-level institutions and firm-level liquidity by documenting that the relationship between liquidity and the SOA is more (less) pronounced for firms operating in weak (strong) institutional environments.

This study contributes to the literature in several ways. It is the first to enrich the literature of leverage adjustment by identifying liquidity as a new determinant of SOA. Second, it provides further evidence that the impact of liquidity on leverage SOA is not homogeneous but distinct for over- and under-levered firms as a result of the different financing behaviors of these firms. It also highlights the important roles of institutional environments in shaping a firm’s capital structure decisions by providing new empirical evidence of the joint effect of liquidity and institutional contexts.

This study has several implications. Firms’ managers may improve equity liquidity at the firm-level to enhance its positive impact on leverage SOA. Moreover, policy makers may use regulations to strengthen their law and order traditions, create good government, enforce creditor rights, and improve the banking sector. Better institutional environments create higher leverage SOA.

Supplemental Material

Supplemental_Material – Supplemental material for Liquidity and speed of leverage adjustment

Supplemental material, Supplemental_Material for Liquidity and speed of leverage adjustment by Ly Ho, Yue Lu and Min Bai in Australian Journal of Management

Footnotes

Final transcript accepted 22 March 2020 by Philip Gharghori (AE Finance).

Author’s note

Ly Ho is also affiliated with School of Accounting, Finance and Economics, Waikato Management Division, University of Waikato, Hamilton, New Zealand.

Funding

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

ORCID iDs

Yue Lu

Min Bai

Supplemental material

Supplemental material for this article is available online.

Notes

References

Amihud

(2002) Illiquidity and stock returns: Cross-section and time-series effects. Journal of Financial Markets 5: 31–56.

Amihud

Mendelson

(1986) Asset pricing and the bid-ask spread. Journal of Financial Economics 17: 223–249.

(2015) Firm crash risk, information environment, and speed of leverage adjustment. Journal of Corporate Finance 31: 132–151.

Bancel

Mittoo

(2004) Cross-country determinants of capital structure choice: A survey of European firms. Financial Management 33: 103–132.

Berkman

Nguyen

(2010) Domestic liquidity and cross-listing in the United States. Journal of Banking & Finance 34: 1139–1151.

Brennan

Subrahmanyam

(1996) Market microstructure and asset pricing: On the compensation for illiquidity in stock returns. Journal of Financial Economics 41: 441–464.

Brennan

Chordia

Subrahmanyam

(1998) Alternative factor specifications, security characteristics, and the cross-section of expected stock returns. Journal of Financial Economics 49: 345–373.

Brockman

Chung

(2003) Investor protection and firm liquidity. The Journal of Finance 58: 921–937.

Butler

Grullon

Weston

(2005) Stock market liquidity and the cost of issuing equity. Journal of Financial and Quantitative Analysis 40: 331–348.

10.

Byoun

(2008) How and when do firms adjust their capital structures toward targets? The Journal of Finance 63: 3069–3096.

11.

Chai

Faff

Gharghori

(2010) New evidence on the relation between stock liquidity and measures of trading activity. International Review of Financial Analysis 19: 181–192.

12.

Chan

Faff

(2003) An investigation into the role of liquidity in asset pricing: Australian evidence. Pacific-Basin Finance Journal 11: 555–572.

13.

Chang

Chou

Huang

(2014) Corporate governance and the dynamics of capital structure: New evidence. Journal of Banking & Finance 48: 374–385.

14.

Cheung

Noe

, et al. (2019) Liquidity, information production, and debt-equity choice. Working Paper. Available at: https://papers.ssrn.com/sol3/Papers.cfm?abstract_id=3426640

15.

Claessens

Djankov

Klapper

(2003) Resolution of corporate distress in East Asia. Journal of Empirical Finance 10: 199–216.

16.

Cook

Tang

(2010) Macroeconomic conditions and capital structure adjustment speed. Journal of Corporate Finance 16: 73–87.

17.

Çolak

Gungoraydinoglu

Öztekin

(2018) Global leverage adjustments, uncertainty, and country institutional strength. Journal of Financial Intermediation 35: 41–56.

18.

Corwin

Schultz

(2012) A simple way to estimate bid-ask spreads from daily high and low prices. The Journal of Finance 67: 719–760.

19.

Cumming

Johan

(2011) Exchange trading rules and stock market liquidity. Journal of Financial Economics 99: 651–671.

20.

Dang

Moshirian

, et al. (2019) News media coverage and corporate leverage adjustments. Journal of Banking & Finance 109: 105666.

21.

Dang

Moshirian

Wee

CKG

, et al. (2015) Cross-listings and liquidity commonality around the world. Journal of Financial Markets 22: 1–26.

22.

Datar

Naik

Radcliffe

(1998) Liquidity and stock returns: An alternative test. Journal of Financial Markets 1: 203–219.

23.

Davydenko

Franks

(2008) Do bankruptcy codes matter? A study of defaults in France, Germany, and the UK. The Journal of Finance 63: 565–608.

24.

Demirgüç-Kunt

Maksimovic

(1996) Stock market development and financing choices of firms. The World Bank Economic Review 10: 341–369.

25.

Demirgüç-Kunt

Maksimovic

(1999) Institutions, financial markets, and firm debt maturity. Journal of Financial Economics 54: 295–336.

26.

Devos

Rahman

Tsang

(2017) Debt covenants and the speed of capital structure adjustment. Journal of Corporate Finance 45: 1–18.

27.

Djankov

Hart

McLiesh

, et al. (2008) Debt enforcement around the world. Journal of Political Economy 116: 1105–1149.

28.

Djankov

La Porta

Lopez-de-Silanes

, et al. (2003) Courts. The Quarterly Journal of Economics 118: 453–517.

29.

Doidge

Karolyi

Stulz

(2007) Why do countries matter so much for corporate governance? Journal of Financial Economics 86: 1–39.

30.

Drobetz

Wanzenried

(2006) What determines the speed of adjustment to the target capital structure? Applied Financial Economics 16: 941–958.

31.

Drobetz

Schilling

Schröder

(2015) Heterogeneity in the speed of capital structure adjustment across countries and over the business cycle. European Financial Management 21: 936–973.

32.

Edmans

Fang

Zur

(2013) The effect of liquidity on governance. The Review of Financial Studies 26: 1443–1482.

33.

Elsas

Florysiak

(2011) Heterogeneity in the speed of adjustment toward target leverage. International Review of Finance 11: 181–211.

34.

Fan

Titman

Twite

(2012) An international comparison of capital structure and debt maturity choices. Journal of Financial and Quantitative Analysis 47: 23–56.

35.

Faulkender

Flannery

Hankins

, et al. (2012) Cash flows and leverage adjustments. Journal of Financial Economics 103: 632–646.

36.

Fischer

Heinkel

Zechner

(1989) Dynamic capital structure choice: Theory and tests. The Journal of Finance 44: 19–40.

37.

Flannery

Rangan

(2006) Partial adjustment toward target capital structures. Journal of Financial Economics 79: 469–506.

38.

Fong

Holden

Trzcinka

(2017) What are the best liquidity proxies for global research? Review of Finance 21: 1355–1401.

39.

Frank

Goyal

(2009) Capital structure decisions: Which factors are reliably important? Financial Management 38: 1–37.

40.

Frieder

Martell

(2006) On capital structure and the liquidity of a firm’s stock. Available at: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=880421

41.

Gao

Zhu

(2015) Information asymmetry and capital structure around the world. Pacific-Basin Finance Journal 32: 131–159.

42.

Goldstein

Leland

(2001) An EBIT-based model of dynamic capital structure. The Journal of Business 74: 483–512.

43.

Goyenko

Holden

Trzcinka

(2009) Do liquidity measures measure liquidity? Journal of Financial Economics 92: 153–181.

44.

Hackbarth

Miao

Morellec

(2006) Capital structure, credit risk, and macroeconomic conditions. Journal of Financial Economics 82: 519–550.

45.

Hasbrouck

(2009) Trading costs and returns for US equities: Estimating effective costs from daily data. The Journal of Finance 64: 1445–1477.

46.

Huang

Ritter

(2009) Testing theories of capital structure and estimating the speed of adjustment. Journal of Financial and Quantitative Analysis 44: 237–271.

47.

Kahn

Winton

(1998) Ownership structure, speculation, and shareholder intervention. The Journal of Finance 53: 99–129.

48.

La Porta

Lopez-de-Silanes

Shleifer

, et al. (1998) Law and finance. Journal of Political Economy 106: 1113–1155.

49.

La Porta

Lopez-de-Silanes

Shleifer

, et al. (2002) Investor protection and corporate valuation. The Journal of Finance 57: 1147–1170.

50.

Leary

Roberts

(2005) Do firms rebalance their capital structures? The Journal of Finance 60: 2575–2619.

51.

Liao

LKC

Mukherjee

Wang

(2015) Corporate governance and capital structure dynamics: An empirical study. Journal of Financial Research 38: 169–192.

52.

Lipson

Mortal

(2009) Liquidity and capital structure. Journal of Financial Markets 12: 611–644.

53.

Liu

(2009) Historical market-to-book in a partial adjustment model of leverage. Journal of Corporate Finance 15: 602–612.

54.

Maug

(1998) Large shareholders as monitors: Is there a trade-off between liquidity and control? The Journal of Finance 53: 65–98.

55.

Modigliani

Miller

(1958) The cost of capital, corporation finance and the theory of investment. The American Economic Review 48: 261–297.

56.

Morck

Yeung

(2000) The information content of stock markets: Why do emerging markets have synchronous stock price movements? Journal of Financial Economics 58: 215–260.

57.

Myers

Majluf

(1984) Corporate financing and investment decisions when firms have information that investors do not have. Journal of Financial Economics 13: 187–221.

58.

Nadarajah

Ali

Liu

, et al. (2018) Stock liquidity, corporate governance and leverage: New panel evidence. Pacific-Basin Finance Journal 50: 216–234.

59.

Noe

(2002) Investor activism and financial market structure. The Review of Financial Studies 15: 289–318.

60.

Öztekin

(2015) Capital structure decisions around the world: Which factors are reliably important? Journal of Financial and Quantitative Analysis 50: 301–323.

61.

Öztekin

Flannery

(2012) Institutional determinants of capital structure adjustment speeds. Journal of Financial Economics 103: 88–112.

62.

Pagan

(1984) Econometric issues in the analysis of regressions with generated regressors. International Economic Review 25: 221–247.

63.

Rajan

Zingales

(1995) What do we know about capital structure? Some evidence from international data. The Journal of Finance 50: 1421–1460.

64.

Shleifer

Wolfenzon

(2002) Investor protection and equity markets. Journal of Financial Economics 66: 3–27.

65.

Stoll

Whaley

(1983) Transaction costs and the small firm effect. Journal of Financial Economics 12: 57–79.

66.

Strebulaev

(2007) Do tests of capital structure theory mean what they say? The Journal of Finance 62: 1747–1787.

67.

Udomsirikul

Jumreornvong

Jiraporn

(2011) Liquidity and capital structure: The case of Thailand. Journal of Multinational Financial Management 21: 106–117.

68.

Warr

Elliott

Koëter-Kant

, et al. (2012) Equity mispricing and leverage adjustment costs. Journal of Financial and Quantitative Analysis 47: 589–616.

69.

Zhou

Tan

KJK

Faff

, et al. (2016) Deviation from target capital structure, cost of equity and speed of adjustment. Journal of Corporate Finance 39: 99–120.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.51 MB