Debt Structure and Financial Flexibility 1

Sean Flynn

Arizona State University

Sean.J.Flynn@asu.edu

September 18, 2016

Abstract

I study the relation between firm debt structure and future financial flexibility. Iconsider how the total level of debt, maturity, security, and priority may potentiallyimpact a firm’s ability to raise new financing and undertake profitable investments. Ifind that firms with lower total debt (high debt capacity) are more financially flex-ible. Lower leverage increases future new debt issues and investment, and firms donot fully rebalance by reducing the use of external financing sources such as equity.Furthermore, in contrast to previous empirical results, I find that greater reliance onlong-term debt may be associated with higher ex-post flexibility, in particular a signif-icantly higher amount of investment. This is consistent with theoretical predictions onrollover risk. Finally, my results support the view that greater reliance on unsecureddebt can increase future debt financing. Overall, my paper offers new insights into howaspects of debt structure are related to financial flexibility.

1Department of Finance, Arizona State University, P.O. Box 873906, Tempe, AZ 85287-3906;sjflynn1@asu.edu, 605-999-5170. I thank Yuri Tserlukevich, Andra Ghent, Mike Hertzel, Luke Stein, andIlona Babenko for helpful comments and discussions thus far. The most recent version of this paper can befound at https://sites.google.com/site/seanjflynnjr/research

1 Introduction

I study the relation between debt structure and future financial flexibility. Recent literature

suggests that firms’ desire to maintain financial flexibility is a key missing component of

capital structure theories. Flexibility is valuable to firms because it allows them to under-

take profitable investment opportunities as they arise, as well as avoid financial distress in

the face of negative shocks to profitability. As suggested by DeAngelo and DeAngelo (2007),

incorporating firms’ desire for flexibility into capital structure models may result in predic-

tions that more closely align with some “puzzling” empirical regularities, such as the fact

that many profitable firms forgo interest tax shields by maintaining very low debt levels.

Theoretical models suggest that various characteristics of debt can affect future financial

flexibility, and empirical evidence shows firms may ex-ante select a particular maturity or

security profile based on a desire for flexibility. For example, Johnson (2003) shows that

firms with greater future growth opportunities select shorter overall debt maturity in order

to mitigate future debt overhang. However, no paper to date has considered how matu-

rity, security, and priority of a firm’s current debt actually facilitate future financing and

investment.

The goal of my paper is to link debt structure and observed financing and investment,

and offer a new understanding of the relation between current debt and future flexibility. I

estimate the relation between current debt structure and future financing and investment,

focusing on the total level of debt, the maturity profile, the mix of secured and unsecured

debt, and the mix of senior and junior/subordinated debt. These debt characteristics are

motivated in large part by the literature on how firms’ debt structure can mitigate agency

costs, particularly debt overhang and the associated underinvestment. Maturity, security,

and priority of current debt directly affect a firm’s ability to finance new investment with

new equity and debt. Therefore, firms may rely on a greater degree of short-term debt or

a greater proportion of unsecured or lower priority debt in their current capital structure if

1

they anticipate costly agency problems.1

I measure flexibility in terms of both new issues of debt and equity and in terms of net

external financing (new issues net of reductions). Furthermore, I exploit large, “proactive”

increases in debt and equity (as identified and defined by Denis and McKeon 2012 and

McKeon 2015) to obtain additional results. These are substantial new financing transactions

that indicate a large amount of flexibility. Furthermore, they can be tied to a particular use

of funds, e.g., acquisitions or increases in internal capital expenditures. As such, they provide

a unique setting for further understanding how debt structure is related to the ability of firms

to finance major investment with new debt and equity.

My results suggest first that the current total level of debt has a strong negative relation

with flexibility. A higher level of total debt is associated with lower future debt financing

and investment. The effects are economically meaningful, with a one standard deviation

increase in market leverage implying lower net debt issuance of over 5%, which is nearly

twice the average net debt issuance. A one standard deviation increase in leverage is also

associated with 15% lower average investment. Furthermore, higher debt is associated with a

lower probability of large, “proactive” increases in debt and equity that are motivated by the

desire to increase long-term investment. This is not simply the result of rebalancing, in which

a firm substitutes equity for debt in order to lower its leverage ratio, because pure rebalancing

would imply no change in total financing and investment. On the contrary, my results show

both lower total financing and investment for firms with high leverage. Additionally, the

results are unlikely to be driven by a mechanical feedback effect from growth options to

leverage, whereby growth options increase the market value of the firm and depress total

leverage (Berens and Cuny 1995, Tserlukevich 2008). I conduct robustness checks using only

the subsample of firms that are likely to have more valuable growth options, and I also use

book leverage instead of market leverage. Both tests indicate that the presence of growth

options cannot fully explain why low total debt is associated with greater future external

1See, e.g., Myers (1977), Barnea et al (1980), and Stulz and Johnson (1985).

2

financing.

I interpret the negative relation between total leverage and future debt financing and

investment as indicating that higher total debt reduces flexibility. This may be due to several

channels. Higher debt may reduce future debt capacity, and hence the ability to finance new

investment with debt, either by increasing the probability of default or by putting a firm

beyond its debt capacity in the sense of Myers and Majluf (1984). Higher debt may also

impose greater debt overhang (Myers 1977), which can lead firms to underinvest in positive

NPV projects. The latter channel would predict lower equity issuance. Because I find future

debt issuance to be more affected than future equity issuance, I conclude that the association

between total debt and flexibility is primarily due to a debt capacity channel, as opposed to

an agency cost channel.

The results further show that short maturity may be associated with lower financial

flexibility, although the effect is less pronounced. A greater proportion of debt maturing in

1-3 years is associated with less net debt financing, a lower probability of large, investment-

motivated debt issues, and lower capital expenditures. Again, the relation is economically

meaningful. A one standard deviation increase in the proportion of short-maturity debt is

associated with 16% lower average net debt issuance and 1.2% lower average investment.

Therefore, short-maturity debt may actually reduce flexibility, rather than enhance it. Al-

though this is consistent with a rollover risk channel, in which short-term debt can reduce

debt capacity ex-ante because of a higher ex-post probability of default,2 it contrasts with

much of the theoretical literature that finds short-maturity debt may enhance future flexibil-

ity by reducing or eliminating the effects of debt overhang on new financing and investment.3

Additionally, my results call into question the interpretation of previous empirical studies

that show that firms ex-ante select into shorter-maturity debt structures when the value of

flexibility increases. For example, Giambona et al (2015) show that an exogenous increase

in growth opportunities results in firms shortening their debt maturity, which they interpret

2See, e.g., He and Xiong (2012) and He and Milbradt (2016).3See, e.g., Myers (1977), Barnea et al (1980), Childs et al (2005), and Titman and Tsyplakov (2007).

3

as firms’ attempts to mitigate debt overhang. While this may be true, my results show that

firms that maintain more short-maturity debt may ultimately have less flexibility ex-post.

Finally, I find that unsecured debt weakly increases future net debt issuance, but has an

ambiguous effect on future investment. The relation with debt issuance is consistent with

the predictions of, e.g., Stulz and Johnson (1985) and Hackbarth and Mauer (2012) who

show that firms can preserve the option to issue new secured or senior debt in the future by

relying on more unsecured or lower priority debt in the present. The ability to issue secured

or more senior debt to finance investment mitigates the effect of debt overhang, allowing

firms to increase their future flexibility.

These results are important because they provide new insights into how aspects of current

debt are related to future flexibility. Most of the existing empirical literature on debt and

flexibility draws conclusions by examining the ex-ante choice of debt characteristics given

firms’ anticipated need for flexibility.4 In contrast, my results show how debt maturity,

security, and priority are related to actual financing and investment. Furthermore, none

of the current studies have documented a negative relation between short-maturity debt

and ex-post flexibility.5 Thus my findings call into question the interpretation of previous

empirical evidence and suggest that, in fact, long-term debt may be associated with greater

flexibility ex-post.

The remainder of this paper is structured as follows. Section 2 discusses literature and

motivates the link between current debt structure and future financing and investment.

Section 3 outlines the data and empirical methodology, Sections 4 and 5 discuss results, and

Section 6 concludes.

4See Barclay and Smith (1995a), Barclay and Smith (1995b), Goyal, Lehn, and Racic (2002), Johnson(2003), Billett et al (2007), and Giambona et al (2015)

5Aivazian, Ge, and Qiu (2005) examine how maturity is related to future investment and find a positiverelation between short-maturity debt and investment. However, their study does not consider the securityand priority of debt, nor do they examine the relation between maturity and future financing.

4

2 The relation between debt structure and financial

flexibility

A large theoretical literature shows that debt structure can improve or detract from flexibility

through various channels. In this paper I consider four aspects of debt structure that are

related, theoretically, to flexibility: (1) the total level of debt, (2) the mix of short- vs long-

maturity debt, (3) the mix of secured vs unsecured debt, and (4) the mix of senior and

junior/subordinated debt.

The literature generally agrees that a higher total level of debt can reduce future financing

and investment through a number of channels. As outlined in Myers and Majluf (1984) and

Myers (1984), a modified pecking order view of capital structure would predict that firms

have a particular “debt capacity” beyond which financing with additional debt becomes very

costly. This would imply that firms that are close to or at capacity are less flexible, all else

equal, than firms that are far away. On the other hand, firms that are far away from their

debt capacity are more able to obtain new financing and exercise growth options. A higher

level of debt may also reduce future debt capacity and flexibility via a standard tradeoff

theory channel of Modigliani and Miller (1963). If, as in Leland (1994), firms trade off the

tax benefits of higher debt with the cost of increased risk of distress or bankruptcy, then a

high level of existing debt may reduce the ability or desire to issue more debt in the future.

Finally, total debt may affect flexibility because existing risky debt can create debt over-

hang (Myers 1977). Equity holders in a firm with a large amount of outstanding debt may

underinvest in positive NPV projects if they anticipate that existing debt holders will reap

a large portion of the gains at their expense. This implies that higher levels of existing debt

can reduce both new equity and debt issuance and thus limit a firm’s ability to exercise

valuable growth options. The recent work of Sunderesan et al (2015) offers further support

for the debt overhang channel in a dynamic setting. They show that optimal leverage is

lower for firms that expect to exercise valuable growth options in the future.

5

From both a debt capacity and agency cost perspective, a higher level of debt is associated

with lower financial flexibility. Various characteristics of debt, however, can enhance or

diminish flexibility conditional on a given level of total debt.

A higher proportion of short-maturity debt can increase flexibility because it provides

a firm more frequent opportunities to roll over or refinance. This in turn creates more

frequent opportunities for a firm to increase its total debt level (by either rolling over the

entire amount and also issuing new debt, or by allowing all debt to mature and issuing more),

or reduce its total debt (by allowing some or all of the debt to mature). A greater ability

to adjust total debt increases a firm’s ability to respond to positive investment shocks or

negative profitability shocks.

Short-maturity debt may also increase financing and investment flexibility by mitigating

or eliminating the effects of debt overhang (see, e.g., Myers 1977, Childs et al 2005, and

Titman and Tsyplakov 2007). This is possible because of the timing of when short-term

debt matures relative to when the firm wants to exercise its growth option. If the short-

term debt matures prior to when the firm wishes to invest, then shareholders can make

the investment decision as if the firm was all equity-financed. They can issue new debt to

fund the investment, and because the new debt will be priced such that the benefits will

not accrue to debt holders, the underinvestment problem is entirely resolved. Even if the

debt matures after the investment is made, short-term debt can at the very least mitigate

underinvestment.

Barnea et al (1980) argue that short-term debt can mitigate a different agency cost of

debt: risk-shifting. If equity holders can benefit from shifting into a higher-risk, lower-return

project at the expense of bond holders, then they may face a higher ex-ante cost of debt, as

bond holders will rationally discount the price at which they are willing to purchase debt.

This would imply lower debt capacity, all else equal. Barnea et al (1980) show that short

maturity can mitigate this investment distortion because the value of short-maturity debt is

less sensitive to an increase in risk than the value of longer-maturity debt, hence bond holders

6

are expropriated less. Leland and Toft (1996) also suggest risk-shifting as an explanation

for the observed reliance on short-term debt. They argue that long-maturity debt allows for

larger debt capacity and higher tax shields, therefore the propensity for firms to use short

term term debt must be explained by the existence of bond holder-stock holder conflict over

investment policy.

Despite the potential for short-maturity debt to increase flexibility by reducing agency

costs of debt, the literature also suggests that greater reliance on short-term debt can expose

a firm to more frequent rollover losses and may thus reduce the ability of firms to access

and restructure financing. For example, the recent work of He and Xiong (2012) and He and

Milbradt (2016) predict that more short-maturity debt can increase the incentive of equity

holders to default early. This is because a larger amount of short-maturity debt means more

frequent refinancing, and if equity holders must absorb refinancing losses (the difference

between the face value of maturing bonds and the proceeds from issuing the new debt), then

they choose to default sooner. This implies lower debt capacity ex-ante.

In addition to maturity, the mix of secured and unsecured, or senior and junior, debt

can also affect flexibility by mitigating the effects of debt overhang on future financing and

investment. Stulz and Johnson (1985) show that the ability of firms to issue new, secured

debt allows them to undertake investment opportunities they would otherwise forgo if they

had to be financed via unsecured debt or equity. This is because the new debt can be secured

by the investment, which limits the ability of existing unsecured debt holders to capture the

benefits. In a dynamic model in which firms can issue debt and invest in multiple periods,

Hackbarth and Mauer (2012) assume firms can prioritize debt issues in a way that minimizes

over- and underinvestment. Like the static results of Stulz and Johnson (1985), their model

predicts that issuing more senior debt today can lead to future underinvestment but can also

mitigate future overinvestment. Thus, a key implication of their model is that the choice

of whether to prioritize the current debt issue or future debt issues (or to make them equal

priority) may impact whether the firm invests in the future.

7

3 Data and empirical methodology

3.1 Data

My primary data sample consists of North American Compustat firms from 1978-2015. I

require that each firm-year have greater than $10 million in assets in order to be included

in the sample, and I exclude observations that are missing any of the explanatory variables

required for my primary tests. I exclude utilities and financial firms.

Table 1 provides variable definitions, and Table 2 shows summary statistics. I consider

four aspects of debt as constituting a firm’s debt structure: the total level, the maturity

structure, the security profile, and the seniority profile. I measure the total level of debt as

market leverage (Mkt lev): book value of debt divided by the sum of book value of debt plus

market value of equity. As an alternative, I also use the book value of leverage (Book lev),

which is equal to book debt divided by book assets.

In line with existing empirical studies, I define maturity structure in terms of the propor-

tion of short-maturity debt. In particular, I define Short−maturity as the ratio of long-term

debt maturing within the next three years to total debt. I construct the numerator by sum-

ming Compustat items DD1, DD2, and DD3 (the proportion of long-term debt maturing in

one, two, and three years, respectively). This measure is identical to the maturity measures

used in Johnson (2003) and Billett et al (2007).6

Finally, I define security and priority structure based on the extent to which firms use

unsecured or junior/subordinated debt in their current capital structure. Consistent with

the previous literature, I define Unsecured as the ratio of unsecured debt to total debt, where

unsecured debt is the difference between total debt and secured debt (Compustat item dm).

I define Subordinated as the ratio of subordinated unsecured debt to total unsecured debt.

Financial flexibility entails the ability to access and restructure external financing at

low cost, and the ability to engage in profitable investment opportunities. Therefore, I

6This variable is the complement to the measure used in Barclay and Smith (1995a) in that they use theproportion of long-term debt maturing in more than three years in the denominator.

8

measure ex-post flexibility in terms of debt and equity financing and investment. My primary

measures of debt financing are new debt issuance (dissue), which is defined as increases in

long-term debt (Compustat item dltis) scaled by lagged total assets, and net debt issuance

(ndissue), which is defined as long-term debt increases net of reductions (item dltis minus

dltr) scaled by lagged total assets. Similarly, my primary measures of equity financing are

new equity issuance (eissue), which is defined as the sale of common and preferred stock

(Compustat item sstk) scaled by lagged total assets, and net equity issuance (neissue),

which is equal to the issue of new stock net of repurchases (item sstk minus prstkc) scaled

by lagged total assets. I define two additional measures that capture total external financing:

net external financing (netexternal), which is equal to the sum of ndissue and neissue, and

new external financing (newexternal), which is equal to the sum of dissue and eissue. The

latter measure captures the extent to which firms engage in new financing, whereas the former

captures the net effect of changes in bond and stock issuance and reductions/repurchases.

Finally, I define investment (investment) as capital expenditures scaled by lagged total

assets.

As an alternative measure of financing and investment, I construct three measures of

large, new debt and equity issues. Large financing choices are, by nature, more indicative

of greater flexibility than the financing choices captured by the variables dissue and eissue.

The ability to engage in a large new debt or equity issue to fund, e.g., a major acquisition,

indicates a high degree of access to external financing. To measure large financing choices

that are used primarily for long-term investment, I follow methodology derived from Denis

and McKeon (2012) to define transactions that I call large, proactive increases in debt or

equity (LPIDs and LPIEs, respectively).7 I also define a second measure of large equity

increases based on McKeon (2015) which I refer to as Sstk3. 8 This variable indicates

whether the firm issued new stock equal to 3% or more of its total equity in a given year.

7Denis and McKeon 2012 only focus on large, proactive increases in debt. However, I use their method-ology to define a symmetric transaction for increases in equity.

8See Appendix for a detailed explanation of how these variables are defined.

9

3.2 Regression model

I construct my main empirical specification to estimate the relation between debt structure

and future financing and investment as defined in the previous subsection. I estimate:

Yi,t = β0 + β1MktLeveragei,t−1 + β2Short Maturityi,t−1 +

+β3Unsecuredi,t−1 + β4Subordinatedi,t−1 + βxConti,t−1 + εi,t (1)

The dependent variable is one of the following: dissue, eissue, newexternal, ndissue,

neissue, netexternal, or investment. The right-hand side variables are one-year lagged

levels of the four debt characteristics of interest. Control variables include firm characteris-

tics such as market-to-book, size, tangibility, dividend paying status, R&D, cash holdings,

and profitability, trade credit, a rating dummy, as well as firm and year fixed effects.

The relation between debt and future flexibility may be more obvious, however, in the

context of large financing and investment choices. Therefore, I also estimate how the debt

structure measures are related to the three types of large financing transactions I discussed

in the previous subsection: LPIDs, LPIEs, and Sstk3:

Largei,t = β0 + β1MktLeveragei,t−1 + β2Short Maturityi,t−1 +

+β3Unsecuredi,t−1 + β4Subordinatedi,t−1 + βxConti,t−1 + εi,t (2)

where the dependent variable is one of three variables: (1) LPIDi,t, which is a dummy equal

to 1 if firm i engages in a large, proactive increase in debt in year t, and 0 otherwise; (2)

LPIEi,t, which is a dummy equal to 1 if firm i engaged in a large, proactive increase in equity

in year t, and 0 otherwise; or (3) Sstk3i,t, which is a dummy equal to 1 if the firm issued

new stock equal to 3% or more of its total equity in a given year, and 0 otherwise (McKeon

2015). The independent variables of interest and controls are the same as in equation 1, and

I use industry-by-year fixed effects. I estimate the equation using a linear probability model.

10

4 Results

4.1 Total level of debt

Table 3 shows results for estimating equation 1 for new debt, new equity, and total new

external financing. A higher level of total debt (row 1) is associated with significantly lower

new debt issues and is not significantly related to new equity issues. The relation with total

new external financing (column 3) is also negative and significant, indicating that firms’

combined equity and debt issuance following high market leverage is lower.

Consistent with the findings in Table 3, Table 4, which show the results from estimating

equation 2 using a linear probability model, illustrates that higher levels of debt are associ-

ated with a significantly lower probability of engaging in large, proactive increases in debt

or equity. The −0.29 coefficient indicates that a one standard deviation increase in market

leverage is associated with a 7.5% lower probability of a LPID, which is large relative to the

unconditional probability of 9.9% and as such is economically meaningful. This negative sign

is consistent with results in Table 3 for new debt issues. However, the negative association

between total debt and large, proactive increases in equity (LPIEs) suggests that, although

higher leverage may have a positive (albeit statistically insignificant) effect on new equity

issuance (column 2 of Table 3), it has a significant negative effect on future large equity

issues.

Finally, Table 5 shows results for estimating equation 1 for net debt issuance, net equity

issuance, total net external financing, and investment. The first row shows that market

leverage (Mkt lev) is significantly and negatively related to future net debt issuance, and

positively related to future net equity issuance, although the effect on equity is much smaller.

The coefficient indicates that a one standard deviation increase in market leverage in the

previous year is associated with a roughly 5% decrease in net debt issuance, whereas a one

standard deviation increase is only associated with a 0.7% increase in net equity. Further-

more, market leverage is also negatively related to both net and new external financing,

11

indicating that combined debt and equity financing is lower following a period of higher

leverage. Finally, total leverage is negatively related to investment, indicating that firms

invest less relative to their average in a period following higher leverage. The coefficient

indicates that a one standard deviation increase in leverage is associated with a decrease

in investment of about 1% relative to average. Given average investment of 6.8%, this is

economically meaningful and equivalent to a decrease of 15% of average investment.

An obvious explanation for the negative relation between current debt and future debt

and equity financing shown in Tables 3-5 is rebalancing: a firm with higher-than-optimal

leverage may reduce its leverage by lower its debt and increasing its equity. Although this is

likely part of the story, the results do not suggest it is the primary driver. Pure rebalancing,

in which the firm substitutes equity for debt, should result it no change in total financing

or investment. The fact that both investment and total financing decrease following a high

level of debt indicates that the result cannot be driven entirely by rebalancing. Furthermore,

the evidence for large debt and equity issues suggests that significant increases in both types

of financing are less likely following a period of high total debt, which would further indicate

that it is not purely rebalancing.

Given that rebalancing is likely not the primary channel, the results are consistent with

the total amount of debt reducing flexibility. In particular, the results suggest that higher

debt levels reduce future flexibility via a debt capacity channel—more debt reduces future

new debt capacity, leading to a lower ability to access new financing. Although there may

also be an effect via the agency cost channel, as higher debt is associated with lower future

investment, the positive relation between past debt and net equity financing in Table 5

makes this less obvious. If investment is lower purely due to debt overhang created by higher

previous levels of debt, then equity issuance should also be negatively affected, because debt

overhang reduces investment through its effect on new equity issuance. If equity holders’

unwillingness to finance new investment with new equity were the driving force behind the

negative total debt-investment relation, then we would not expect a positive relation between

12

past total debt and future equity. Although Table 4 does show that total debt is associated

with a lower probability of a new large equity issue, the impact on the probability of a

large debt issue is larger in magnitude than the impact on the probability of a large equity

issue. Thus, the results in Tables 5 and 4 support that higher total debt reduces flexibility

specifically through a debt capacity channel.

One possible alternative interpretation is that higher previous levels of debt are related

to lower new debt issuance due to the mechanical effect of growth options. Because options

contribute positively to the market value of the firm, they result in lower leverage ratios

(see, e.g., Berens and Cuny 1995 and Tserlukevich 2008). Therefore, it may appear that

the low-leverage firms raise more external financing and also invest more in the future. To

address this possibility, I estimate equations 1 and 2 using book leverage (Book lev) in place

of market leverage. Book leverage should be less sensitive to growth options, because the

denominator is the book value, not the market value, of assets. The results for new financing,

shown in Table 6, and for net financing and investment, shown in Table 7, are qualitatively

unchanged when book leverage is used.

As an additional check, I reestimate equations 1 and 2 on subsamples of firms that likely

have more valuable options. In particular, I measure yearly sales growth and define firms

above the median to be “high growth” firms.9 Alternatively, I group firms above the median

market-to-book ratio. If the negative total debt and financing/investment relation is driven

primarily by the presence of growth options, then the effect should become insignificant in

a sample of firms that all have low growth opportunities. For both definitions of growth,

however, I find that higher total debt is still associated with lower future debt financing and

investment for low-growth firms, as shown in Tables 8 and 9. Although the magnitude of

the coefficients is smaller than for the subsamples of high-growth firms, the effects are still

statistically significant, indicating that the feedback effect cannot be the primary driver of

the results.

9Billett et al (2007) advocate the use of sales growth as a measure of growth opportunities that is analternative to market-to-book ratio.

13

4.2 Maturity

Table 3 illustrates that the primary measure of short-term debt, the proportion of total debt

maturing within 1-3 years (Short−maturity), is associated with greater new debt financing

(column 1) and total new external financing (column 3). This would appear at first to suggest

that greater reliance on short-maturity debt may increase flexibility as it increases the amount

of future new debt financing. However, the results in Table 5 suggest that short-term debt

may actually reduce flexibility. Column 1 shows that Short −maturity is associated with

lower net debt issuance, indicating that although firms issue more new debt, the simultaneous

reductions of existing debt are larger in magnitude. The coefficient indicates that a one

standard deviation increase in the proportion of debt maturing in 1-3 years is associated

with lower net debt issuance of 0.4%, which translates into an economically meaningful

decrease of 16% of average net debt issuance. There is no statistically significant relation

with net equity issuance and thus the effect on combined net debt and equity financing is

negative and significant (column 3). Furthermore, column 4 indicates that more short-term

debt is associated with less future investment—the coefficient implies a decrease of 1.2% of

average investment for a one standard deviation increase in Short−maturity

The view that short-maturity debt may actually reduce flexibility is further supported

by Table 4, which shows that Short−maturity is negatively associated with the probability

of a large, proactive increase in debt (LPID). The coefficient indicates that a one standard

deviation increase in the proportion of debt maturing in 1-3 years is associated with a 0.7%

lower probability of engaging in a LPID, which is economically meaningful given that the

unconditional probability of a LPID is 9.9%. Although Short−maturity is positively related

to a large, proactive increase in equity (LPIE), the magnitude of the coefficient is much

smaller.

The results in Table 4 are important for two reasons. First, the debt and equity trans-

actions used on the left-hand side are tied primarily to increases in long-term investment—

acquisitions or capital expenditures—as opposed to other uses of funds. Thus, the strong,

14

negative relation in column 1 and the much weaker, but positive, relation in column 2 further

supports the findings in Table 5 that short-maturity debt may actually reduce overall exter-

nal financing and investment. Second, the LPID and LPIE transactions are major financing

choices that indicate substantial financial flexibility. Although there is no way to “net” the

LPIDs and LPIEs, the relative sizes of the coefficients on Short − maturity in columns 1

and 2 suggest that greater reliance on short-maturity debt lowers the likelihood of a large

financing transaction, and thus decreases flexibility, “on net.”

One interpretation of the negative sign between short-maturity debt and future net debt

issuance (column 1 of Table 5) is the following: firms with shorter-maturity debt are more

able to reduce their level of debt in the face of a negative shock to profitability, which

allows them to avoid financial distress (see, e.g., Dangl and Zechner 2016). Firms hit with

a negative shock optimally respond by lowering their net debt issuance regardless, but a

greater proportion of short-term debt should allow a firm to do this more quickly. If this

is the case, then the negative relation between short maturity and net debt issuance may

actually indicate that short maturity enhances flexibility, rather than diminishes it.

In order to understand whether the negative relation between short-maturity and net

debt issuance is driven by firms that are optimally reducing their debt in response to being

close to distress, I split the sample into firms with negative profitability and firms with zero

or positive profitability. Firms with negative profitability should value the ability to quickly

reduce debt more than firms with nonnegative profitability, so I would expect the association

between maturity and net debt issuance to be stronger in the sample of negative profitability

firms, if in fact the debt reduction motive is driving the results. Table 10 shows the results

of reestimating equation 1 on these two subsamples. The negative short maturity-net debt

relation remains significant in the subsample of firms with nonnegative profitability. This

further suggests that the negative maturity-net debt relation is a sign of reduced, as opposed

to increased, flexibility.

Although not as robust as the results for the total level of debt, the maturity results con-

15

trast with the previous empirical evidence on the maturity-flexibility relation. Much of the

theoretical literature finds that short-maturity debt enhances future flexibility by reducing

or eliminating the effects of debt overhang on new financing and investment. Consistent with

these predictions, Barclay and Smith (1995a) show that firms with higher growth opportu-

nities have more short-term debt, and Johnson (2003) and Billett et al (2007) show that

short maturity may enhance flexibility by reducing the negative relation between growth

options and leverage. Goyal, Lehn, and Racic (2002) and Giambona et al (2015) provide

evidence that firms choose ex-ante shorter (longer) maturity when flexibility becomes more

(less) valuable as measured by changes in future growth options. While it may be true that

firms ex-ante select more short-maturity debt given a greater need for flexibility, my results

suggest that, ex-post, firms’ overall debt issuance and ability to engage in large, new debt

financing to fund investment may not be higher. Because firms do not appear to fully sub-

stitute this shortfall with equity, overall, more short-term debt may be associated with lower

flexibility.

Overall, the results suggest a negative relation between short-maturity debt and future

investment that contrasts with theoretical predictions regarding how short-maturity debt

can mitigate debt overhang. The negative relation is consistent, however, with the rollover

channel described in the recent work of He and Xiong (2012) and He and Milbradt (2016).

Their models predict that more short-maturity debt can increase the incentive of equity

holders to default early. This is because a larger amount of short-maturity debt means more

frequent refinancing, and if equity holders must absorb refinancing losses (the difference

between the face value of maturing bonds and the proceeds from issuing the new debt), then

they choose to default sooner. This implies lower debt capacity, and hence reduced external

financing ability, ex-ante.

16

4.3 Security and priority

Table 3 illustrates that greater reliance on unsecured debt is not significantly associated with

new debt or equity financing. However, table 5 shows that a greater proportion of unsecured

debt is associated with greater net debt issuance. The coefficient on unsecured debt implies

an increase of roughly 16% of average net debt issuance. This finding is consistent with

the predictions of Stulz and Johnson (1985) and Hackbarth and Mauer (2012), who show

that firms can “preserve priority” for future debt in a way that facilitates future issues and

minimizes over- and underinvestment. More precisely, consistent with my results, relying

primarily on unsecured debt provides an option to issue both more unsecured and new

senior debt in the future, making effective costs smaller and debt issues more frequent.

Additionally, because existing risky debt creates debt overhang, secured debt allows firms to

undertake investment opportunities they would otherwise forgo if they had to be financed

via unsecured debt or equity. Finally, unsecured debt also encourages new debt financing

because issuing more senior debt dilutes the value of the claims of existing debt holders.

The resulting agency conflict however, has a potential to distort investment and can result

in lower firm value.

Despite this, greater reliance on unsecured debt is also associated with somewhat lower

net equity issuance and lower future investment. Additionally, it is not significantly related

to the probability of a large, proactive debt or equity increase (Table 4). Thus, although

the result for net debt financing is consistent with firms being able to generate future debt

capacity by preserving security for future debt issues, I cannot conclude that unsecured debt

has a strong or robust relation with future flexibility.

Similarly, the priority structure of unsecured debt also does not appear to be robustly

related to future financing and investment. Tables 3 and 5 illustrate that having more

subordinated debt is associated with higher equity financing, but Subordinated is statistically

insignificant for the debt financing and investment variables.

17

5 Robustness

The full-sample analysis in Section 4 suggests a strong negative relation between total debt

and flexibility, a weaker negative relation between short-maturity debt and flexibility, and

a weaker positive relation between unsecured debt and flexibility. The following section

further investigates how these aspects of debt are related to future financing and investment

and discusses how the presence of selection bias can affect the interpretation of the results.

I conduct two robustness checks. First, I consider whether debt characteristics are more

strongly related to future flexibility within subsamples of firms that should benefit more

from access to external financing and investment. Second, I specify an alternative investment

equation to further establish the robustness of the negative relation between short-maturity

and investment.

5.1 Subsample analysis

Although the unsecured and subordinated debt measures are not particularly powerful or

consistent in the full sample, they may have greater explanatory power in subsamples of firms

for which flexibility is more valuable. To test whether debt structure may increase flexibility

for certain types of firms more that others, I estimate equation 1 on three subsamples of firms

for which flexibility should be more valuable: (1) higher-growth firms, (2) lower-profitability

firms, and (3) financially constrained firms.

Higher-growth firms should value flexibility more because more of their firm value is

embedded in growth options. Therefore, the cost of forgone investment is higher for these

firms. Low-profitability firms should value flexibility because, all else equal, they should

be closer to financial distress than higher-profitability firms. Financially constrained firms

should value flexibility because they have lower ex-ante access to external sources of financing.

Given the greater value these types of firms should place on financial flexibility, I predict

that the relation between debt structure and future financing and investment will be stronger

18

(i.e., the coefficients will be larger in magnitude) for these three subsamples of firms.

I define the subsamples as follows. For high-growth firms, I take the set of firms with

yearly sales growth above the median. As an alternative, I use firms with greater-than-

median market-to-book ratio. For low-profitability firms, I consider firms with negative

profitability. Alternatively, I take the set of firms below the median profitability. For fi-

nancially constrained firms, I compute the financial constraint index in Hadlock and Pierce

(2010), which is a function of firm size and age. In particular, the index is constructed as

SAi,t = (−.737 ∗ sizei,t) + .042 ∗ (size2i,t) + (−.040 ∗ agei,t)

where size is the log of book assets and age is the number of years for which the firm has

appeared in Compustat relative to year t. For example, if the firm has appeared since 2000,

and t = 2005, then age is equal to 5. A higher value of the index indicates a more constrained

firm. I use firms above the median value of the SA index as the constrained subsample.

5.1.1 High- vs low-growth firms

Tables 8-12 show the results. I only include results for total net issuance, total new external

financing, and investment. In Table 8, I compare high- vs low-growth firms, where growth is

defined in terms of yearly sales growth. Firms above the median sales growth are considered

high-growth firms. The effect of total leverage on net debt issuance and investment is stronger

for the subsample of high-growth firms, consistent with costlier agency conflicts, and hence

more valuable flexibility, for these firms. The primary measure of short maturity, as shown

in Table 8, continues to be significantly associated with lower investment, and the effect is

stronger in the high-growth subsample. Finally, the measure of unsecured debt with respect

to net debt issuance loses significance in the high-growth sample, although it remains positive

and significant in the low-growth subsample.

As an alternative to sales growth, I also partition the sample by market-to-book ratio

19

(MTB), and consider firms above the median to be high-growth firms. Table 9 shows the

results. Consistent with results in Table 8, total leverage has a greater effect on future net

debt issuance and investment in the subsample of high market-to-book firms. The relation

between unsecured debt and net debt issuance is positive and significant in both high- and

low-MTB subsamples.

5.1.2 High- vs low-profitability firms

In Table 10, I compare the relation between debt structure and flexibility for firms with

negative profitability against firms with zero or positive profitability. Unlike in the high-

vs low-growth subsamples, the differential impact of total leverage is less pronounced. All

else equal, negative profitability firms should benefit more from reducing leverage, as they

are closer to financial distress. However, the coefficient magnitudes for negative profitability

firms are somewhat lower than the magnitudes for positive profitability firms. For the set of

negative profitability firms, the only debt characteristic that remains significant is the total

level of debt. However, in the set of firms with zero or positive profitability, the coefficient

signs on the maturity and security measures are consistent with those in Table 5.

As an alternative, I cut the sample by median profitability and report the results in

Table 11. Consistent with the previous results for negative vs positive profitability, the

impact of total debt is somewhat stronger for the high-profitability firms. The negative

relation between maturity and net debt issuance and investment remains significant in the

subsample of firms above median profitability but loses significance in the low-profitability

sample.

5.1.3 Constrained vs unconstrained firms

Table 12 shows results for the subsample of constrained vs unconstrained firms, where con-

strained firms are defined as having a value of SA greater than the median. Consistent with

smaller, younger firms benefiting more from lower leverage, the coefficient magnitudes for the

20

effects of market leverage on net external financing, new external financing, and investment

are larger in the sample of constrained firms. The associations between maturity and new

external financing and investment are again mixed.

Overall, the analysis of these three subsamples of firms is broadly consistent with the

full sample analysis. The total level of debt appears to be more important for high-growth

and constrained (younger and/or smaller) firms. The differences in coefficients for short-

maturity debt across subsamples generally do not suggest that firms that value flexibility

more necessarily benefit more from longer maturity. Unsecured debt generally has a greater

potential impact on future net debt issuance for lower-growth firms, but consistent with

constrained firms benefiting more from preserving future debt priority, it has a large impact

on net debt issuance for constrained firms.

5.2 Alternative investment model

Two previous empirical studies, Aivazian, Ge, and Qiu (2005) and Dang (2011), find that

more long-term debt is associated with lower future investment, a result that seems to run

against the result in my paper. In order to check the robustness of my investment results,

I estimate an alternative specification of the investment equation. In particular, I use the

investment-cash flow sensitivity model employed in both Aivazian, Ge, and Qiu (2005) and

Dang (2011). This model is dynamic in nature and includes as explanatory variables lagged

investment and a measure of cash flow, as well as measures of growth opportunities and

total leverage. In line with Aivazian, Ge, and Qiu (2005) and Dang (2011) I also include my

measure of maturity. Finally, I include my other two measures of debt structure:

Investmenti,t = β0 + β1MktLeveragei,t−1 + β2Short Maturityi,t−1 +

+β3Unsecuredi,t−1 + β4Subordinatedi,t−1 + βxConti,t−1 + εi,t (3)

21

The controls include lagged investment, cash flows (cflow, the sum of operating income

before extraordinary items and depreciation, divided by lagged book assets), market-to-

book, and firm and year fixed effects.10 The inclusion of firm fixed effects accounts for the

fact that lagged investment will be correlated with an unobservable (“ommitted”) firm effect.

As Aivazian, Ge, and Qiu (2005) and Dang (2011) point out, equation 3 uses lagged

investment as an explanatory variable, which will be correlated with an individual firm

effect. The inclusion of firm fixed effects controls for the individual effect, but the coefficient

estimates are still biased because the within-firm mean of lagged investment is correlated

with the mean of the error term. I thus first-difference equation 3 as follows:

∆Investmenti,t = β0 + β1∆MktLeveragei,t−1 + β2∆Short Maturityi,t−1 +

+β3∆Unsecuredi,t−1 + β4∆Subordinatedi,t−1 + βx∆Conti,t−1 + εi,t (4)

The dependent variable is the first difference of investment, and the independent variables are

lagged first differences. This transformation removes the time-invariant firm effect. Following

Dang (2011), I then instrument the lagged first-difference of investment with the second-

lagged level of investment. That is, I include Investmenti,t−2 in place of ∆Investmenti,t−1

on the right-hand side of equation 4.

I estimate equation 4 and report the results in Table 13. The negative relation between

Short−maturity and investment is robust to this alternative specification.

It is possible that the negative maturity-investment relation is driven by a common,

unobservable factor: firms with shorter maturity debt invest less for reasons not related

to maturity. Although I cannot entirely rule this out, the use of firm-year fixed effects in

equations 1 and 3 partially controls for this concern.

10To maintain consistency with Aivazian, Ge, and Qiu (2005), I winsorize the cash flow variable cflowsuch that the maximum and minimum values are 5 and -5.

22

5.3 Causality and Selection

My empirical strategy does not allow me to cleanly separate a causal effect of debt structure

from a selection effect. A selection effect would imply that firms select into a particular

debt structure ex-ante based on their expectations about the need to fund investment with

external financing in the future. If, for example, firms anticipate profitable investment

opportunities in the future, they may choose a particular level of debt and/or maturity

and security profile that will allow them to undertake the investment. This may drive the

observed relation between debt structure and future flexibility and would bias the regression

coefficients.

Despite the potential for selection bias, it is unlikely that it is the main driver of my

results. First, it is difficult to believe that firms can anticipate investment, and hence the

need for financing, so far in advance. Consistent with this conjecture, my results remain

significant when I separate the sample by ex-ante measures of growth options, as I discuss

in Section 5.1.1. Additionally, such an explanation appears inconsistent with the observed

ex-post differences in equity and debt financing.

Finally, the results on maturity in particular are informative regardless of the extent

to which selection drives the results. If selection is the primary driver, then my result

suggests firms select into longer-maturity debt in anticipation of future need for financing

and investment. This would run contrary to both existing empirical evidence and many

theoretical predictions regarding how firms select short-maturity debt in order to mitigate

underinvestment. On the other hand, if the regression coefficients primarily pick up a causal

effect of debt structure on future flexibility, then my results suggest that short-maturity debt

does not actually increase future debt financing capacity or mitigate debt overhang.

23

6 Conclusion

I study the link between current debt structure and future financial flexibility by estimating

the relation between total debt, maturity, security, and priority and future financing and

investment. My results show that the total level of debt may reduce flexibility through a

negative effect on future external financing and investment. The negative relation is not

simply the result of rebalancing, nor is it due to a mechanical feedback effect from growth

options to leverage. Furthermore, I do not find that total debt negatively impacts future

flexibility primarily via an agency cost channel, whereby higher debt imposes greater debt

overhang. I conclude that the effect is primarily a result of higher leverage reducing future

debt capacity. The results further suggest that, contrary to existing empirical work and

many theoretical predictions, a greater proportion of short-term debt may be associated with

lower ex-post flexibility. Despite the maturity-flexibility relation being generally less robust,

it nevertheless is consistent with short-term debt creating rollover risk that can effectively

reduce ex-ante debt capacity. Finally, I find evidence that unsecured debt increases future

debt issuance, but has an ambiguous effect on future investment.

24

7 References

Aivazian, V., Ge, Y., and J. Qiu. (2005). “Debt Maturity Structure and Firm Investment.”

Financial Management, 34, 107-119.

Barclay, M. and C. Smith. (1995a). “The maturity structure of corporate debt.” Journal

of Finance, 50, 609-631.

Barclay, M. and C. Smith. (1995b). “The priority structure of corporate liabilities.”

Journal of Finance, 50, 899-916.

Barnea, A., Haugen, R., and L. Senbet. (1980). “A rationale for debt maturity structure

and call provisions in the agency theoretic framework.” Journal of Finance, 35, 1223-1234.

Berens, J. and C. Cuny. (1995). “The Capital Structure Puzzle Revisited.” Review of

Financial Studies, 8, 1185-1208.

Billet, M., King, T., and D. Mauer. (2007). “Growth opportunities and the choice of

leverage, debt maturity, and covenants.” Journal of Finance, 62, 697-730.

Childs, P., Mauer, D. and S. Ott. (2005). “Interactions of corporate financing and

investment decisions: The effects of agency conflicts.” Journal of Financial Economics, 76,

667-690.

Dang, V. (2011). “Leverage, Debt Maturity, and Firm Investment: An Empirical Anal-

ysis.” Journal of Business, Finance, and Accounting, 38, 225-258.

Dangl, T. and J. Zechner. (2016). “Debt Maturity and the Dynamics of Leverage.”

Working paper, http : //papers.ssrn.com/sol3/papers.cfm?abstractid = 890228.

DeAngelo, H. and L. DeAngelo. (2007). “Capital structure, payout policy, and finan-

cial flexibility.” Working paper, http : //papers.ssrn.com/sol3/papers.cfm?abstractid =

916093 Denis, D. and S. McKeon. (2012). “Debt financing and financial flexibility: Evi-

dence from proactive leverage increases.” Review of Financial Studies, 25, 1897-1929.

Gamba, A. and A. Triantis. (2008). “The value of financial flexibility.” Journal of

Finance, 63, 2263-2296.

Giambona, E., Golec, J. and F. Lopez-de-Silanes. (2015). “Financing Choices and

25

Growth Prospects: Evidence from the Introduction of Biosimilar Drugs.” Working paper,

http : //papers.ssrn.com/sol3/papers.cfm?abstractid = 2162376

Goyal, V., Lehn, K., and S. Racic. (2002). “Growth opportunities and corporate debt

policy: the case of the U.S. defense industry.” Journal of Financial Economics, 64, 35-59.

Hackbarth, D. and D. Mauer. (2012). “Optimal priority structure, capital structure, and

investment.” Review of Financial Studies, 25, 747-796.

Hadlock, C. and J. Pierce. (2010). “New Evidence on Measuring Financial Constraints:

Moving Beyond the KZ Index.” Review of Financial Studies, 23, 1909-1940.

He, Z. and W. Xiong. (2012). “Rollover Risk and Credit Risk.” Journal of Finance, 67,

391-429.

He, Z. and K. Milbradt. (2016). “Dynamic Debt Maturity.” Review of Financial Studies,

forthcoming. Johnson, S. (2003). “Debt maturity and the effects of growth opportunities

and liquidity risk on leverage.” Review of Financial Studies, 16, 209-236.

Leland, H. (1994). “Corporate debt value, bond covenants, and optimal capital struc-

ture.” Journal of Finance, 49, 1213-1252.

Leland, H. and K. Toft. (1996). “Optimal capital structure, endogenous bankruptcy,

and the term structure of credit spreads.” Journal of Finance, 51, 987-1019.

McKeon, S. (2015). “Employee Option Exercise and Equity Issuance Motives.” Working

paper, http : //papers.ssrn.com/sol3/papers.cfm?abstractid = 1920985.

Modigliani, F. and M. Miller. (1963). “Corporate income taxes and the cost of capital:

A correction.” American Economic Review, 53, 433-443.

Myers, S. (1977). “Determinants of Corporate Borrowing.” Journal of Financial Eco-

nomics, 5, 147-175.

Myers, S. (1984). “The Capital Structure Puzzle.” Journal of Finance, 39, 575-592.

Myers, S. and N. Majluf. (1984). “Corporate financing and investment decisions when

firms have information that investors do not have.” Journal of Financial Economics, 13,

187-221.

26

Stulz, R. and H. Johnson. (1985). “An analysis of secured debt.” Journal of Financial

Economics, 14, 501-521.

Sundaresan, S., Wang, N., and J. Yang. (2015). “Dynamic investment, capital structure,

and debt overhang.” Review of Corporate Finance Studies, 4, 1-42.

Titman, S. and S. Tsyplakov. (2007). “A dynamic model of optimal capital structure.”

Review of Finance, 11, 401-451.

Tserlukevich, Y. (2008). “Can real options explain financing behavior?” Journal of

Financial Economics, 89, 232-252.

27

A Appendix

A.1 Identifying large, proactive increases in debt

In order to identify large, proactive increases in debt (LPIDs), I utilize the following method,

which I derive from Denis and McKeon (2012).

First, I identify firm-years in which (1) total debt increased relative to the previous year

and (2) market leverage increased by more than 0.1 relative to the previous year. That

is, firm-years in which the year-over-year change in market leverage was greater than 10

percentage points. Market leverage is equal to the book value of debt divided by the sum of

the book value of debt and market capitalization. Because market leverage is bound between

0 and 1, this equates to an increase of 10 percentage points or more. This is the way in which

I define “large.”

Second, from that subset of observations, I further identify which increases in leverage

are predominantly a result of a large increase in debt, as opposed to an exogenous decline in

equity value. Here, I follow Denis and McKeon (2012) and define a variable $∆MLit which

captures the value of additional market leverage resulting from the increase:

$∆MLit = Di,t −Di,t−1MAit

MAi,t−1

(5)

where Dit is the total debt of firm i in year t and MAit is the sum of market value of equity

and book value of debt. In order to screen out increases in leverage that result from declines

in equity value, I require that the increase in total debt I observe is at least 90% of $∆MLit.

In other words, if total debt increases by 9, and $∆MLit = 10, then I include it.

As an example of how this measure allows for the screening out of increases in leverage

that result from declines in equity value, assume the market value of assets in year t − 1 is

100, and total debt is 20, meaning the firm’s market leverage is 0.20. If the firm issues an

additional 30 in debt in year t with no change in the value of equity, its assets increase to

28

130, and its market leverage increases to 0.38. The value of additional debt resulting from

the increase, $∆MLit, is 24. Therefore, the increase in total debt is 125% of $∆MLit, well

in excess of the 90% threshold. Now assume that, instead of issuing additional debt in year

t, the firm keeps debt at 20, but the market value of equity drops to 32.6 in year t. Assets

therefore decrease to 52.63, but leverage increases (due to the decrease in equity) to 0.38.

So, we have the same change in market leverage, except there has been no change in debt.

In this case, $∆MLit = 9.47. But since debt did not change, the 90% threshold is not met,

so this observation would be screened out.

This gives me a subset of large leverage increases that are primarily the result of a firm

taking on more debt. Finally, from this set of observations I select only those for which I can

identify, using the Statement of Cash Flows (SCF), that the funds from the debt issuance

were primarily associated with (1) an increase in long-term investment (capital expenditures

or acquisitions), (2) engaging in a payout to shareholders or repurchasing stock, (3) an

increase in working capital, or (4) covering an operational cash flow shortfall from, e.g., a

negative earnings shock.11 In particular, following Denis and McKeon (2012) I require that

the combined uses of funds from the SCF comprise at least 80% of the observed increase in

debt. For example, if a firm’s debt increases from $100 million in year t to $200 million in

year t+ 1, the screen requires that data on at least $80 million of the three potential uses of

funds be available.

In addition to filtering the sample based on the SCF, I use the breakdown of the use

of funds to classify each LPID into a primary use of funds category. That is, I categorize

the LPIDs as primarily being used for long-term investment, payouts, increases in working

capital, covering operational cash shortfalls, or for “multiple” uses. The primary use of funds

is defined by whichever category comprises greater than 50% of the total percentage of debt

increase captured on the SCF. For example, if debt increases $100 and the SCF captures

that entire $100 increase, and $75 of the $100 captured on the SCF is used for acquisitions,

11See the appendix for a more detailed description of how each use of funds is defined and derived fromthe SCF.

29

then the primary use of funds is classified as long-term investment.

As an example of the SCF screen, in 2002, the market leverage of Lee Enterprises, a

publishing company, doubled from 11% (in 2001) to 22% as a result of an increase of $235

million in total debt, which corresponded to an increase in net debt12 of $494 million. Based

on the SCF, the proceeds from the issue were used for an increase in long-term investment

of $675 million. Therefore, the use of funds comprises 136% of the increase in net debt, well

in excess of the 80% threshold I require. (Note that percentages greater than 100% are not

uncommon, as firms often have additional sources of funds, such as equity, that they use to

fund operational needs). Additionally, because the percentage of the increase in net debt

attributable to long-term investment was greater than the percentage attributable to the

other three potential uses, the screening process flags long-term investment as the primary

use of funds. A reading of the firm’s 10 − K for that year confirms that it took on the

additional debt in order to help finance the acquisition of Howard Publication Co.13

A.2 Identifying large, proactive increases in equity

I identify large increases in equity using two methods. First, I define large, proactive increases

in equity (LPIEs) in a manner analogous to the definition of LPIDs described above. The

only difference is the following. Because I am interested in increases in equity that result in

a significant change in leverage, I require that leverage decreased by more than 0.1 relative

to the previous year. This is simply the opposite of the criterion for large debt increases.

Furthermore, I measure changes in book leverage, as opposed to market leverage. That is,

I require the following two criteria: (1) total book equity increased relative to the previous

12Net debt is equal to total debt minus cash holdings. See Gamba and Triantis (2008) for a discussion ofhow debt issuance costs can lead firms to hold cash despite having outstanding debt.

13As another example of the increases I identify, in 2013, the market leverage of MDC Holdings Coincreased to 42% from 31% in 2012. This was the result of an increase in total debt of $337.5 million, whichcorresponded to an increase in net debt of $298 million. Based on the SCF, the proceeds from the issuewere used for an increase in long-term investment of $60 million, as well as an increase in working capitalof $233 million. Therefore, the use of funds comprises 98% of the increase in net debt, well in excess of the80% threshold I require. Additionally, because the percentage of the increase in net debt attributable toincreases in working capital was greater than the percentage attributable to the other three potential uses,the screening process flags working capital as the primary use of funds.

30

year, and (2) book leverage decreased by more than 0.1 relative to the previous year.

As an alternative method of identifying large equity increases, I use the definition of

McKeon (2015), which measures whether the firm issued new stock equal to 3% or more

of its total equity in a given year (“firm-initiated issues” as opposed to“employee-initiated

issues”).

31

Tab

le1:

Var

iable

defi

nit

ions

Var

iable

Des

crip

tion

Mkt

lev

Tot

aldeb

tdiv

ided

by

the

sum

ofto

tal

deb

tan

dm

arke

tca

pit

aliz

atio

nB

ook

lev

Tot

aldeb

tdiv

ided

by

the

book

valu

eof

asse

tsU

nse

cure

dU

nse

cure

ddeb

tdiv

ided

by

tota

ldeb

tS

hort

-mat

uri

tyD

ebt

mat

uri

ng

in3

year

sor

less

div

ided

by

tota

ldeb

tS

ubo

rdin

ated

Pro

por

tion

ofunse

cure

ddeb

tth

atis

sub

ordin

ated

LP

IDD

um

my

equal

to1

iffirm

-yea

rhas

incr

ease

indeb

tdefi

ned

inA

pp

endix

A.1

LP

IED

um

my

equal

to1

iffirm

-yea

rhas

incr

ease

ineq

uit

ydefi

ned

inA

pp

endix

A.2

Sst

k3D

um

my

equal

to1

iffirm

-yea

rhas

new

equit

yis

suan

ceth

atex

ceed

s3%

ofen

d-o

f-p

erio

deq

uit

ydi

ssu

eIn

crea

ses

inlo

ng-

term

deb

tdiv

ided

by

lagg

edas

sets

eiss

ue

Sal

eof

com

mon

and

pre

ferr

edst

ock

div

ided

by

lagg

edas

sets

ndi

ssu

eIn

crea

ses

inlo

ng-

term

deb

tnet

ofre

duct

ions,

div

ided

by

lagg

edas

sets

nei

ssu

eSal

eof

com

mon

and

pre

ferr

edst

ock

net

ofre

purc

has

es,

div

ided

by

lagg

edas

sets

new

exte

rnal

Incr

ease

sin

long-

term

deb

tplu

ssa

leof

com

mon

and

pre

ferr

edst

ock

,div

ided

by

lagg

edas

sets

net

exte

rnal

Sum

ofndis

sue

and

nei

ssue

inve

stm

ent

Cap

exdiv

ided

by

asse

tsca

shC

ash

hol

din

gsdiv

ided

by

asse

tsrd

R&

Dex

pen

sediv

ided

by

asse

tssi

zeL

og(a

sset

s)m

tb(A

sset

s-

shar

ehol

der

seq

uit

y-

def

erre

dta

xes

+m

arke

tca

p+

pre

fst

ock

liquid

atin

gva

lue)

/as

sets

profi

tabi

lity

Op

erat

ing

inco

me

bef

ore

dep

reci

atio

ndiv

ided

by

asse

tsta

ngi

bili

tyN

etpro

per

ty,

pla

nts

,an

deq

uip

men

tdiv

ided

by

asse

tsdi

vide

nd

paye

rD

um

my

equal

to1

ifth

efirm

pai

da

div

iden

din

the

give

nye

arap

ayA

ccou

nts

pay

able

div

ided

by

tota

las

sets

rtg

dum

my

Dum

my

equal

to1

ifth

efirm

-yea

rhas

aS&

Pcr

edit

rati

ng

cash

flow

Sum

ofop

erat

ing

inco

me

bef

ore

extr

aord

inar

yit

ems

and

dep

reci

atio

n,

div

ided

by

lagg

edb

ook

asse

tssa

legr

Yea

r-ov

er-y

ear

per

centa

gech

ange

insa

les

grow

thS

AH

adlo

ckan

dP

ierc

e(2

010)

finan

cial

const

rain

tin

dex

age

The

num

ber

ofye

ars

for

whic

hth

efirm

has

app

eare

din

Com

pust

at

32

Table 2: Summary Statistics

Variable Obs. Mean Std. Dev. Min MaxMkt lev 133099 0.257 0.255 0 0.94Book lev 150427 0.266 0.246 0 1Unsecured 112588 0.624 0.374 0 1Short-maturity 132494 0.343 0.333 0 1Subordinated 130963 0.049 0.152 0 0.828LPID 128609 0.099 0.299 0 1LPIE 150369 0.08 0.271 0 1Sstk3 146549 0.265 0.442 0 1dissue 144168 0.098 0.183 0 1eissue 146705 0.067 0.175 0 0.974ndissue 127622 0.025 0.146 -0.295 0.832neissue 124646 0.061 0.252 -0.161 1.75newexternal 128310 0.226 0.459 0 2.837netexternal 117750 0.092 0.323 -0.239 2.097investment 148218 0.068 0.076 0 0.508size 151037 5.223 1.9 2.303 10.461mtb 128028 1.833 1.727 0.46 47.219profitability 150089 0.076 0.205 -4.72 0.432tangibility 150742 0.299 0.237 0 0.945dividend payer 151037 0.387 0.487 0 1rd 83388 0.079 0.122 0 0.692cash 140174 0.111 0.157 0 0.893apay 150952 0.091 0.079 0.003 0.426rtg dummy 151037 0.196 0.397 0 1cflow 134976 0.185 1.688 -5 5salegr 134190 1.203 0.618 0.283 5.369SA 150427 -2.934 0.684 -4.713 -1.475age 150427 9.572 8.272 0 37

Notes: 1) Data is for nonfinancial, nonutility North American Compustat firms from1978-2015. I require that each firm-year have greater than $10 million in assets in order tobe included in the sample, and I exclude observations that are missing any of theexplanatory variables required for my primary tests. 2). All variables are defined in Table1. 3). All variables are winsorized at the 1% level, with the exception of cflow, which iswinsorized with cutoffs at -5 and 5 to maintain consistency with the definition in Aivazian,Ge, and Qiu (2005).

33

Table 3: New financing and investment

New debt New equity Total newMkt lev -0.069*** 0.0062 -0.072***

(0.012) (0.0093) (0.015)Short-maturity 0.017*** 0.0016 0.018***

(0.0044) (0.0044) (0.0061)Unsecured -0.0081 -0.0026 -0.0083

(0.0052) (0.0047) (0.0069)Subordinated 0.018 0.026** 0.046**

(0.017) (0.011) (0.019)Size -0.015*** -0.058*** -0.071***

(0.0033) (0.0036) (0.0049)MTB 0.011*** 0.047*** 0.057***

(0.0016) (0.0028) (0.0030)Profitability -0.017 -0.20*** -0.22***

(0.014) (0.025) (0.025)Tangibility 0.084*** 0.015 0.10***

(0.020) (0.022) (0.029)Dividend payer 0.0079 0.0049 0.013**

(0.0050) (0.0036) (0.0061)R&D 0.00095 0.51*** 0.50***

(0.025) (0.052) (0.057)Cash -0.10*** -0.040* -0.14***

(0.012) (0.021) (0.024)Trade credit 0.21*** 0.14*** 0.33***

(0.045) (0.041) (0.062)Rating dummy -0.033*** 0.020*** -0.011

(0.0065) (0.0041) (0.0077)Constant 0.092*** 0.24*** 0.32***

(0.023) (0.021) (0.033)

Observations 41,590 42,548 40,740R-squared 0.422 0.514 0.487Firm & Yr FE Yes Yes Yes

Notes: 1) Results of estimating linear regressions of new debt issuance (dissue, column 1),new equity issuance (neissue, column 2), and total new external financing (newexternal,column 3) on debt characteristics and controls. 2) All right-hand side variables areone-period lagged levels. 3) Data is for nonfinancial, nonutility North American Compustatfirms from 1978-2015. I require that each firm-year have greater than $10 million in assetsin order to be included in the sample, and I exclude observations that are missing any of theexplanatory variables required for my primary tests. 4) All variables are defined in Table 1.5). ∗ ∗ ∗p < 0.01, ∗ ∗ p < 0.05, and ∗p < 0.1. Standard errors clustered at the firm level.

34

Table 4: Large debt and equity increases

LPID LPIE Sstk3Mkt lev -0.29*** -0.11*** -0.088***

(0.0083) (0.0078) (0.015)Short-maturity -0.022*** 0.0084* 0.0012

(0.0050) (0.0050) (0.0077)Unsecured 0.0025 -0.0053 0.0068

(0.0050) (0.0047) (0.0077)Subordinated -0.016 0.019 0.017

(0.013) (0.012) (0.022)Size -0.0059*** -0.012*** 0.0031

(0.0014) (0.0013) (0.0025)MTB 0.0040** 0.019*** 0.074***

(0.0017) (0.0018) (0.0035)Profitability 0.033*** 0.15*** -0.062***

(0.012) (0.013) (0.023)Tangibility 0.0023 0.011 -0.024

(0.012) (0.011) (0.021)Dividend payer 0.015*** -0.00058 -0.078***

(0.0042) (0.0041) (0.0070)R&D -0.034* 0.19*** 0.43***

(0.019) (0.023) (0.041)Cash -0.15*** 0.073*** -0.0023

(0.013) (0.015) (0.024)Trade credit 0.21*** 0.096*** 0.038

(0.025) (0.025) (0.041)Rating dummy -0.016*** -0.0029 0.038***

(0.0055) (0.0054) (0.011)Constant 0.22*** 0.12*** 0.12***

(0.011) (0.0098) (0.018)Observations 43,154 43,665 42,548R-squared 0.089 0.080 0.183Ind-Yr FE Yes Yes Yes

Notes: 1) Results of estimating a linear regression of the dummy variables LPID, LPIE,and Sstk3 on debt characteristics and controls. 2) All right-hand side variables areone-period lagged levels. 3) Data is for nonfinancial, nonutility North American Compustatfirms from 1978-2015. I require that each firm-year have greater than $10 million in assetsin order to be included in the sample, and I exclude observations that are missing any of theexplanatory variables required for my primary tests. 4) All variables are defined in Table 1.5). ∗ ∗ ∗p < 0.01, ∗ ∗ p < 0.05, and ∗p < 0.1. Standard errors clustered at the firm level.

35

Table 5: Net financing and investment

Net debt Net equity Total net InvestmentMkt lev -0.20*** 0.027*** -0.17*** -0.047***

(0.0082) (0.0086) (0.011) (0.0026)Short-maturity -0.012*** 0.0020 -0.010** -0.0027***

(0.0030) (0.0041) (0.0047) (0.00095)Unsecured 0.012*** -0.0082* 0.0063 -0.0022**

(0.0034) (0.0042) (0.0051) (0.0011)Subordinated -0.011 0.025*** 0.014 0.0040

(0.012) (0.0094) (0.014) (0.0032)Size -0.011*** -0.049*** -0.054*** -0.0022***

(0.0021) (0.0033) (0.0037) (0.00074)MTB 0.0055*** 0.039*** 0.037*** 0.0046***

(0.0013) (0.0028) (0.0027) (0.00042)Profitability -0.021* -0.16*** -0.16*** 0.030***

(0.011) (0.024) (0.024) (0.0031)Tangibility 0.051*** 0.0033 0.055** 0.0053

(0.014) (0.020) (0.023) (0.0071)Dividend payer 0.012*** 0.0030 0.014*** 0.00087

(0.0032) (0.0034) (0.0044) (0.0012)R&D 0.012 0.42*** 0.28*** -0.0016

(0.018) (0.049) (0.048) (0.0051)Cash -0.051*** -0.041** -0.078*** 0.016***

(0.0093) (0.020) (0.021) (0.0034)Trade credit 0.098*** 0.12*** 0.23*** 0.022***

(0.023) (0.036) (0.041) (0.0082)Rating dummy -0.021*** 0.015*** -0.0048 -0.00043

(0.0047) (0.0039) (0.0056) (0.0015)Constant 0.075*** 0.22*** 0.27*** 0.092***

(0.017) (0.024) (0.032) (0.0071)

Observations 40,801 39,286 36,794 43,317R-squared 0.269 0.529 0.495 0.605Firm & Yr FE Yes Yes Yes Yes

Notes: 1) Results of estimating linear regressions of net debt issuance (ndissue, column 1),net equity issuance (neissue, column 2), total net external financing (netexternal, column3), and investment (investment, column 4) on debt characteristics and controls. 2) Allright-hand side variables are one-period lagged levels. 3) Data is for nonfinancial,nonutility North American Compustat firms from 1978-2015. I require that each firm-yearhave greater than $10 million in assets in order to be included in the sample, and I excludeobservations that are missing any of the explanatory variables required for my primarytests. 4) All variables are defined in Table 1. 5). ∗ ∗ ∗p < 0.01, ∗ ∗ p < 0.05, and ∗p < 0.1.Standard errors clustered at the firm level.

36

Table 6: New financing: book leverage

New debt New equity Total new LPID LPIE Sstk3Book lev -0.038*** 0.048*** 0.00099 -0.29*** -0.084*** -0.0078

(0.013) (0.013) (0.024) (0.0084) (0.0084) (0.018)Short-maturity 0.017*** 0.0050 0.022*** -0.033*** 0.0067 0.0031

(0.0045) (0.0044) (0.0082) (0.0051) (0.0051) (0.0078)Unsecured -0.0080 -0.0032 -0.019* 0.0058 -0.0036 0.0078

(0.0052) (0.0047) (0.010) (0.0050) (0.0047) (0.0077)Subordinated 0.015 0.021* 0.027 -0.013 0.014 0.0019

(0.017) (0.011) (0.026) (0.013) (0.012) (0.023)Size -0.016*** -0.058*** -0.090*** -0.0071*** -0.012*** 0.0038

(0.0033) (0.0036) (0.0068) (0.0014) (0.0014) (0.0025)MTB 0.014*** 0.047*** 0.067*** 0.019*** 0.024*** 0.077***

(0.0016) (0.0027) (0.0035) (0.0017) (0.0017) (0.0033)Profitability -0.0077 -0.20*** -0.25*** 0.049*** 0.15*** -0.021

(0.014) (0.025) (0.030) (0.012) (0.013) (0.023)Tangibility 0.080*** 0.0090 0.10*** 0.0069 0.011 -0.028

(0.020) (0.022) (0.038) (0.012) (0.011) (0.021)Dividend payer 0.0094* 0.0057 0.020** 0.017*** 0.0015 -0.072***

(0.0050) (0.0036) (0.0089) (0.0042) (0.0041) (0.0069)R&D 0.0091 0.51*** 0.53*** -0.025 0.22*** 0.65***

(0.025) (0.052) (0.068) (0.022) (0.025) (0.044)Cash -0.098*** -0.033 -0.15*** -0.16*** 0.076*** 0.0011

(0.012) (0.021) (0.029) (0.013) (0.015) (0.024)Trade credit 0.20*** 0.14*** 0.45*** 0.17*** 0.084*** 0.051

(0.045) (0.041) (0.089) (0.027) (0.026) (0.044)Rating dummy -0.034*** 0.017*** -0.028*** -0.0097* -0.0028 0.033***

(0.0065) (0.0042) (0.010) (0.0056) (0.0054) (0.011)Constant 0.085*** 0.23*** 0.37*** 0.20*** 0.11*** 0.078***

(0.023) (0.021) (0.042) (0.010) (0.0097) (0.018)Observations 41,590 42,548 40,740 43,149 43,661 42,544R-squared 0.421 0.514 0.486 0.089 0.078 0.187Firm & Yr FE Yes Yes Yes No No NoInd-Yr FE No No No Yes Yes Yes

Notes: 1) Results of estimating linear regressions of new debt issuance (dissue, column 1),new equity issuance (neissue, column 2), and total new external financing (newexternal,column 3), LPID, LPIE, and Sstk3 on debt characteristics and controls. 2) Allright-hand side variables are one-period lagged levels. 3) Data is for nonfinancial,nonutility North American Compustat firms from 1978-2015. I require that each firm-yearhave greater than $10 million in assets in order to be included in the sample, and I excludeobservations that are missing any of the explanatory variables required for my primarytests. 4) All variables are defined in Table 1. 5). ∗ ∗ ∗p < 0.01, ∗ ∗ p < 0.05, and ∗p < 0.1.Standard errors clustered at the firm level.

37

Table 7: Net financing and investment: book leverage

Net debt Net equity Total net InvestmentBook lev -0.21*** 0.062*** -0.17*** -0.035***

(0.0097) (0.012) (0.016) (0.0027)Short-maturity -0.020*** 0.0048 -0.013** -0.0034***

(0.0031) (0.0041) (0.0058) (0.00097)Unsecured 0.015*** -0.0086** 0.0081 -0.0019*

(0.0035) (0.0042) (0.0062) (0.0011)Subordinated -0.0051 0.021** 0.015 0.0036

(0.012) (0.0095) (0.017) (0.0033)Size -0.013*** -0.049*** -0.068*** -0.0027***

(0.0021) (0.0033) (0.0046) (0.00075)MTB 0.013*** 0.039*** 0.058*** 0.0063***

(0.0013) (0.0027) (0.0030) (0.00042)Profitability -0.012 -0.18*** -0.22*** 0.034***

(0.011) (0.024) (0.027) (0.0030)Tangibility 0.056*** 0.0013 0.071** 0.0042

(0.015) (0.020) (0.028) (0.0071)Dividend payer 0.014*** 0.0041 0.017*** 0.0018

(0.0032) (0.0034) (0.0053) (0.0012)R&D 0.016 0.44*** 0.48*** -0.0047

(0.021) (0.055) (0.062) (0.0064)Cash -0.055*** -0.035* -0.10*** 0.017***

(0.0093) (0.020) (0.026) (0.0034)Trade credit 0.11*** 0.13*** 0.27*** 0.025***

(0.024) (0.038) (0.051) (0.0091)Rating dummy -0.016*** 0.012*** -0.0039 -0.00024

(0.0047) (0.0041) (0.0073) (0.0015)Constant 0.074*** 0.21*** 0.31*** 0.089***

(0.019) (0.024) (0.040) (0.0074)

Observations 40,596 39,083 36,593 43,112R-squared 0.273 0.527 0.491 0.603Firm & Yr FE Yes Yes Yes Yes

Notes: 1) Results of estimating linear regressions of net debt issuance (ndissue, column 1),net equity issuance (neissue, column 2), total net external financing (netexternal, column3), and investment (investment, column 4) on debt characteristics and controls. 2) Allright-hand side variables are one-period lagged levels. 3) Data is for nonfinancial,nonutility North American Compustat firms from 1978-2015. I require that each firm-yearhave greater than $10 million in assets in order to be included in the sample, and I excludeobservations that are missing any of the explanatory variables required for my primarytests. 4) All variables are defined in Table 1. 5). ∗ ∗ ∗p < 0.01, ∗ ∗ p < 0.05, and ∗p < 0.1.Standard errors clustered at the firm level.

38

Table 8: High vs Low sales growth firms

High growth Low growthNet ext New ext Inv Net ext New ext Inv

Mkt lev -0.28*** -0.13*** -0.049*** -0.11*** 0.016 -0.037***(0.028) (0.039) (0.0046) (0.015) (0.026) (0.0029)

Short-maturity -0.0022 0.023* -0.0030* -0.015** 0.013 -0.0020*(0.011) (0.014) (0.0016) (0.0063) (0.010) (0.0012)

Unsecured 0.011 -0.0072 -0.0032* 0.0072 -0.025* -0.0013(0.012) (0.016) (0.0017) (0.0072) (0.013) (0.0014)

Subordinated 0.024 0.055 0.0060 0.0052 -0.0027 0.0022(0.030) (0.040) (0.0045) (0.019) (0.035) (0.0041)

Size -0.097*** -0.12*** -0.0018 -0.029*** -0.045*** -0.00066(0.0081) (0.010) (0.0011) (0.0054) (0.0092) (0.00098)

MTB 0.048*** 0.058*** 0.0047*** 0.033*** 0.051*** 0.0029***(0.0049) (0.0053) (0.00063) (0.0055) (0.0065) (0.00057)

Profitability -0.19*** -0.19*** 0.026*** -0.22*** -0.25*** 0.035***(0.044) (0.047) (0.0049) (0.038) (0.046) (0.0040)

Tangibility 0.12** 0.17*** 0.0081 0.065* 0.11** 0.016*(0.050) (0.062) (0.011) (0.035) (0.049) (0.0084)

Dividend payer 0.014 0.019 -0.00036 0.014*** 0.015 0.0013(0.011) (0.015) (0.0019) (0.0049) (0.0094) (0.0013)

R&D 0.32*** 0.47*** -0.015 0.50*** 0.56*** 0.0053(0.11) (0.11) (0.0096) (0.100) (0.11) (0.0097)

Cash -0.14*** -0.22*** 0.020*** -0.12*** -0.14*** 0.0093*(0.042) (0.046) (0.0051) (0.037) (0.041) (0.0047)

Trade credit 0.42*** 0.55*** 0.023 0.20*** 0.39*** 0.035***(0.094) (0.13) (0.014) (0.056) (0.12) (0.011)

Rating dummy 0.0024 -0.016 0.00030 -0.0018 -0.015 0.00071(0.014) (0.018) (0.0022) (0.0077) (0.013) (0.0016)

Observations 17,193 19,462 20,404 19,147 20,959 22,384R-squared 0.539 0.534 0.708 0.604 0.584 0.613Firm-Yr FE Yes Yes Yes Yes Yes Yes

Notes: 1) Results of estimating linear regressions of total net external financing(netexternal), new external financing (newexternal), and investment (investment) ondebt characteristics and controls. The high-growth sample is defined as firm-years in whichsales growth is above the median, and the low-growth sample is firm-years in which salesgrowth is at or below the median. 2) All right-hand side variables are one-period laggedlevels. 3) Data is for nonfinancial, nonutility North American Compustat firms from1978-2015. I require that each firm-year have greater than $10 million in assets in order tobe included in the sample, and I exclude observations that are missing any of theexplanatory variables required for my primary tests. 4) All variables are defined in Table 1.5). ∗ ∗ ∗p < 0.01, ∗ ∗ p < 0.05, and ∗p < 0.1. Standard errors clustered at the firm level.

39

Table 9: High vs Low MTB growth firms

High MTB Low MTBNet Ext New Ext Investment Net Ext New Ext Investment

Mkt lev -0.29*** -0.11** -0.039*** -0.13*** 0.0046 -0.037***(0.033) (0.048) (0.0049) (0.014) (0.027) (0.0033)

Short-maturity -0.00082 0.019 -0.0011 -0.016** 0.022* -0.0032**(0.0094) (0.012) (0.0013) (0.0063) (0.013) (0.0015)

Unsecured 0.000015 -0.0090 -0.0029* 0.024*** -0.025* 0.00039(0.011) (0.014) (0.0016) (0.0066) (0.015) (0.0016)

Subordinated 0.0047 0.014 -0.0014 -0.0030 0.0073 0.0056(0.031) (0.044) (0.0039) (0.022) (0.032) (0.0052)

Size -0.092*** -0.12*** -0.0013 -0.036*** -0.048*** -0.0025**(0.0075) (0.0094) (0.0011) (0.0051) (0.012) (0.0012)

MTB 0.043*** 0.056*** 0.0037*** 0.033 0.11*** 0.0095***(0.0040) (0.0045) (0.00050) (0.021) (0.032) (0.0031)

Profitability -0.28*** -0.29*** 0.022*** -0.063 -0.086 0.052***(0.037) (0.038) (0.0035) (0.051) (0.067) (0.0059)

Tangibility 0.14*** 0.18*** 0.0055 0.055* 0.093* 0.0061(0.048) (0.058) (0.011) (0.033) (0.053) (0.011)

Dividend payer 0.012 0.017 0.000066 0.011** 0.0093 0.00089(0.0098) (0.015) (0.0020) (0.0052) (0.011) (0.0014)

R&D 0.43*** 0.53*** -0.011 -0.0084 -0.23 -0.0059(0.075) (0.079) (0.0074) (0.13) (0.18) (0.017)

Cash -0.096*** -0.16*** 0.016*** -0.14*** -0.21*** 0.017***(0.037) (0.039) (0.0041) (0.028) (0.039) (0.0062)

Trade credit 0.37*** 0.45*** 0.029** 0.16*** 0.36** 0.022(0.094) (0.12) (0.013) (0.051) (0.15) (0.014)

Rating dummy 0.0092 0.0053 -0.00015 -0.017** -0.050*** -0.00064(0.012) (0.016) (0.0022) (0.0078) (0.015) (0.0020)

Observations 20,203 23,072 24,292 16,376 17,654 18,807R-squared 0.557 0.527 0.673 0.435 0.570 0.639Firm-Yr FE Yes Yes Yes Yes Yes Yes

Notes: 1) Results of estimating linear regressions of total net external financing(netexternal), new external financing (newexternal), and investment (investment) ondebt characteristics and controls. The high MTB sample is defined as firm-years in whichmarket-to-book ratio is above the median, and the low-growth sample is firm-years inwhich market-to-book ratio is at or below the median. 2) All right-hand side variables areone-period lagged levels. 3) Data is for nonfinancial, nonutility North American Compustatfirms from 1978-2015. I require that each firm-year have greater than $10 million in assetsin order to be included in the sample, and I exclude observations that are missing any of theexplanatory variables required for my primary tests. 4) All variables are defined in Table 1.5). ∗ ∗ ∗p < 0.01, ∗ ∗ p < 0.05, and ∗p < 0.1. Standard errors clustered at the firm level.

40

Table 10: Negative vs Nonnegative Profitability firms

Negative profitability Positive profitabilityNet ext New ext Inv Net ext New ext Inv

Mkt lev -0.19*** -0.11* -0.029*** -0.19*** -0.075*** -0.046***(0.050) (0.063) (0.0062) (0.013) (0.022) (0.0030)

Short-maturity -0.0068 0.0041 -0.0013 -0.0056 0.026*** -0.0025**(0.023) (0.023) (0.0025) (0.0051) (0.0081) (0.0011)

Unsecured -0.021 -0.013 -0.0042 0.014** -0.014 -0.00092(0.025) (0.026) (0.0027) (0.0058) (0.011) (0.0013)

Subordinated 0.044 0.12 0.0042 0.0061 0.018 0.0026(0.10) (0.12) (0.0062) (0.016) (0.025) (0.0034)

Size -0.16*** -0.21*** 0.0030 -0.055*** -0.068*** -0.0030***(0.018) (0.019) (0.0021) (0.0044) (0.0074) (0.00085)

MTB 0.057*** 0.066*** 0.0023*** 0.027*** 0.038*** 0.0065***(0.0068) (0.0066) (0.00065) (0.0034) (0.0043) (0.00070)

Profitability -0.27*** -0.32*** 0.0100** -0.039 0.045 0.047***(0.061) (0.059) (0.0042) (0.033) (0.046) (0.0073)

Tangibility 0.14 0.15 -0.044** 0.014 0.049 0.014*(0.11) (0.11) (0.018) (0.025) (0.039) (0.0073)

Dividend payer 0.0020 -0.0065 0.0011 0.013** 0.012 0.00075(0.028) (0.033) (0.0037) (0.0051) (0.0090) (0.0014)

R&D 0.48*** 0.50*** -0.0072 0.029 -0.047 -0.020(0.10) (0.10) (0.0090) (0.067) (0.095) (0.013)

Cash -0.053 -0.063 0.016*** -0.14*** -0.27*** 0.011**(0.053) (0.053) (0.0054) (0.023) (0.031) (0.0048)

Trade credit 0.47*** 0.50*** 0.018 0.27*** 0.51*** 0.034***(0.16) (0.19) (0.018) (0.048) (0.11) (0.012)

Rating dummy 0.0011 0.0047 -0.0026 -0.015** -0.036*** 0.00022(0.050) (0.053) (0.0059) (0.0072) (0.010) (0.0015)

Observations 7,268 8,937 9,271 29,325 31,803 33,841R-squared 0.624 0.596 0.619 0.375 0.463 0.640Firm-Yr FE Yes Yes Yes Yes Yes Yes

Notes:1) Results of estimating linear regressions of total net external financing(netexternal), new external financing (newexternal), and investment (investment) ondebt characteristics and controls. The negative profitability sample is defined as firm-yearsin which profitability is less than 0, and the positive profitability sample is firm-years inwhich profitability is at or above 0. 2) All right-hand side variables are one-period laggedlevels. 3) Data is for nonfinancial, nonutility North American Compustat firms from1978-2015. I require that each firm-year have greater than $10 million in assets in order tobe included in the sample, and I exclude observations that are missing any of theexplanatory variables required for my primary tests. 4) All variables are defined in Table 1.5). ∗ ∗ ∗p < 0.01, ∗ ∗ p < 0.05, and ∗p < 0.1. Standard errors clustered at the firm level.

41

Table 11: High vs Low Profitability firms

Low profitability High profitabilityNet Ext New Ext Investment Net Ext New Ext Investment

Mkt lev -0.17*** -0.068** -0.039*** -0.21*** -0.095*** -0.051***(0.020) (0.031) (0.0034) (0.021) (0.031) (0.0044)

Short-maturity -0.0097 0.042*** -0.0015 0.0057 0.012 -0.0024*(0.0097) (0.011) (0.0015) (0.0071) (0.012) (0.0013)

Unsecured -0.0023 -0.031** -0.0041*** 0.013* 0.0036 0.00086(0.011) (0.015) (0.0016) (0.0080) (0.013) (0.0017)

Subordinated 0.032 0.047 0.0047 -0.0013 0.017 0.0017(0.028) (0.041) (0.0046) (0.023) (0.035) (0.0044)

Size -0.086*** -0.12*** -0.000058 -0.052*** -0.064*** -0.0041***(0.0080) (0.011) (0.0011) (0.0061) (0.0087) (0.0013)

MTB 0.058*** 0.070*** 0.0032*** 0.024*** 0.029*** 0.0069***(0.0050) (0.0052) (0.00054) (0.0043) (0.0054) (0.00087)

Profitability -0.26*** -0.31*** 0.018*** -0.15*** -0.027 0.036***(0.041) (0.042) (0.0036) (0.055) (0.072) (0.012)

Tangibility 0.11** 0.10* -0.015 0.019 0.043 0.024**(0.046) (0.056) (0.011) (0.033) (0.050) (0.0096)

Dividend payer 0.012 0.014 0.0011 0.014* 0.015 0.00050(0.0087) (0.014) (0.0016) (0.0077) (0.013) (0.0019)

R&D 0.46*** 0.50*** -0.011 0.088 0.16 -0.020(0.077) (0.080) (0.0073) (0.10) (0.14) (0.020)

Cash -0.10*** -0.12*** 0.020*** -0.15*** -0.32*** 0.0020(0.035) (0.038) (0.0042) (0.035) (0.045) (0.0067)

Trade credit 0.25*** 0.47*** 0.0093 0.35*** 0.47*** 0.054***(0.079) (0.13) (0.011) (0.068) (0.10) (0.018)

Rating dummy 0.0025 0.00056 -0.00028 -0.011 -0.040*** -0.00029(0.014) (0.019) (0.0023) (0.0092) (0.013) (0.0020)

Observations 18,747 21,458 22,581 17,846 19,282 20,531R-squared 0.576 0.553 0.576 0.433 0.509 0.693Firm-Yr FE Yes Yes Yes Yes Yes Yes

Notes: 1) Results of estimating linear regressions of total net external financing(netexternal), new external financing (newexternal), and investment (investment) on debtcharacteristics and controls. The high profitability sample is defined as firm-years in whichprofitability is above the median, and the low profitability sample is firm-years in whichprofitability is at or below the median. 2) All right-hand side variables are one-periodlagged levels. 3) Data is for nonfinancial, nonutility North American Compustat firms from1978-2015. I require that each firm-year have greater than $10 million in assets in order tobe included in the sample, and I exclude observations that are missing any of theexplanatory variables required for my primary tests. 4) All variables are defined in Table 1.5). ∗ ∗ ∗p < 0.01, ∗ ∗ p < 0.05, and ∗p < 0.1. Standard errors clustered at the firm level.

42

Table 12: Constrained vs unconstrained firms

Constrained UnconstrainedNet ext New ext Inv Net ext New ext Inv

Mkt lev -0.23*** -0.10** -0.045*** -0.15*** -0.034 -0.038***(0.026) (0.042) (0.0044) (0.014) (0.025) (0.0030)

Short-maturity -0.017 0.034*** -0.0026** -0.0042 0.0031 -0.0027(0.012) (0.0085) (0.0010) (0.0054) (0.015) (0.0017)

Unsecured 0.011 -0.024 -0.0034** 0.010 0.00028 -0.0019(0.012) (0.018) (0.0017) (0.0070) (0.013) (0.0015)

Subordinated 0.024 0.069 0.0043 0.026* 0.00072 -0.0024(0.044) (0.060) (0.0060) (0.015) (0.026) (0.0033)

Size -0.14*** -0.17*** -0.0013 -0.041*** -0.058*** -0.0032***(0.011) (0.014) (0.0014) (0.0052) (0.0098) (0.0012)

MTB 0.055*** 0.067*** 0.0040*** 0.023*** 0.039*** 0.0042***(0.0049) (0.0054) (0.00057) (0.0049) (0.0062) (0.00067)

Profitability -0.16*** -0.19*** 0.017*** -0.14*** -0.12** 0.063***(0.036) (0.039) (0.0034) (0.043) (0.060) (0.0071)

Tangibility 0.17*** 0.25*** -0.061*** 0.032 0.049 0.031***(0.057) (0.069) (0.012) (0.026) (0.044) (0.0094)

Dividend payer 0.0041 0.0074 0.0064*** 0.010** 0.00062 -0.0023(0.011) (0.017) (0.0021) (0.0052) (0.0098) (0.0016)

R&D 0.47*** 0.56*** -0.0067 0.47*** 0.48*** -0.017(0.089) (0.091) (0.0081) (0.100) (0.12) (0.012)

Cash -0.070* -0.11*** 0.014*** -0.14*** -0.22*** 0.011**(0.040) (0.041) (0.0048) (0.030) (0.039) (0.0050)

Trade credit 0.43*** 0.51*** 0.024** 0.12** 0.34** 0.048***(0.089) (0.12) (0.012) (0.055) (0.15) (0.015)

Rating dummy -0.070** -0.10** -0.0078 -0.019** -0.031*** 0.0018(0.035) (0.049) (0.0054) (0.0077) (0.011) (0.0015)

Observations 17,260 19,618 20,434 19,333 21,122 22,678R-squared 0.588 0.572 0.652 0.359 0.466 0.665Firm-Yr FE Yes Yes Yes Yes Yes Yes

Notes: 1) Results of estimating linear regressions of total net external financing(netexternal), new external financing (newexternal), and investment (investment) ondebt characteristics and controls. The constrained sample is defined as firm-years in whichthe variable SA is above the median, and the unconstrained sample is firm-years in whichSA is at or below the median. SA is derived from Hadlock and Pierce (2010). 2) Allright-hand side variables are one-period lagged levels. 3) Data is for nonfinancial,nonutility North American Compustat firms from 1978-2015. I require that each firm-yearhave greater than $10 million in assets in order to be included in the sample, and I excludeobservations that are missing any of the explanatory variables required for my primarytests. 4) All variables are defined in Table 1. 5). ∗ ∗ ∗p < 0.01, ∗ ∗ p < 0.05, and ∗p < 0.1.Standard errors clustered at the firm level.

43

Table 13: Alternative investment specification

∆Investmentt ∆Investmentt∆Mkt levt−1 -0.057***

(0.0021)∆Book levt−1 -0.042***

(0.0023)∆Short−maturityt−1 -0.0036*** -0.0042***

(0.00086) (0.00087)∆Unsecuredt−1 -0.0024*** -0.0020**

(0.00093) (0.00094)∆Subordinatedt−1 0.0067*** 0.0055**

(0.0025) (0.0026)∆Cashflowt−1 -0.00091*** -0.00051***

(0.00016) (0.00016)∆MTBt−1 0.0050*** 0.0080***

(0.00047) (0.00055)Investmentt−2 0.076*** 0.081***

(0.0092) (0.0093)

Observations 66,423 66,423R-squared 0.074 0.074Year FE Yes Yes

Notes: 1) Results of estimating a linear regressions of changes in Investment(Investmentt − Investmentt−1) on changes in debt characteristics and controls. 2) Allright-hand side variables are one-period lagged changes (i.e., the difference in levels betweent− 1 and t− 2), with the exception of Investmentt−2, which is the two-period lagged levelof investment. 3) Data is for nonfinancial, nonutility North American Compustat firmsfrom 1978-2015. I require that each firm-year have greater than $10 million in assets inorder to be included in the sample, and I exclude observations that are missing any of theexplanatory variables required for my primary tests. 4) All variables are defined in Table 1.5). ∗ ∗ ∗p < 0.01, ∗ ∗ p < 0.05, and ∗p < 0.1. Standard errors clustered at the firm level.

44