Show Summary Details

Page of

Printed from Oxford Research Encyclopedias, Economics and Finance. Under the terms of the licence agreement, an individual user may print out a single article for personal use (for details see Privacy Policy and Legal Notice).

date: 07 May 2021

# The Pecking Order Theory of Capital Structure

• Murray Z. Frank, Murray Z. FrankCarlson School of Management, University of Minnesota; Shanghai Advanced Institute of Finance
• Vidhan GoyalVidhan GoyalInstitute for Emerging Market Studies, Hong Kong University of Science and Technology
•  and Tao ShenTao ShenSchool of Economics and Management, Tsinghua University

### Summary

The pecking order theory of corporate capital structure developed by states that issuing securities is subject to an adverse selection problem. Managers endowed with private information have incentives to issue overpriced risky securities. But they also understand that issuing such securities will result in a negative price reaction because rational investors, who are at an information disadvantage, will discount the prices of any risky securities the firm issues. Consequently, firms follow a pecking order: use internal resources when possible; if internal funds are inadequate, obtain external debt; external equity is the last resort.

Large firms rely significantly on internal finance to meet their needs. External net debt issues finance the minor deficits that remain. Equity is not a significant source of financing for large firms. By contrast, small firms lack sufficient internal resources and obtain external finance. Although much of it is equity, there are substantial issues of debt by small firms.

Firms are sorted into three portfolios based on whether they have a surplus or a deficit. About 15% of firm-year observations are in the surplus group. Firms primarily use surpluses to pay down debt. About 56% of firm-year observations are in the balance group. These firms generate internal cash flows that are just about enough to meet their investment and dividend needs. They issue debt, which is just enough to meet their debt repayments. They are relatively inactive in equity markets. About 29% of firm-year observations are in the deficit group. Deficits arise because of a combination of negative profitability and significant investments in both real and financial assets. Some financing patterns in the data are consistent with a pecking order: firms with moderate deficits favor debt issues; firms with very high deficits rely much more on equity than debt. Others are not: many equity-issuing firms do not seem to have entirely used up the debt capacity; some with a surplus issue equity. The theory suggests a sharp discontinuity in financing methods between surplus firms and deficit firms, and another at debt capacity. The literature provides little support for the predicted threshold effects.

The theoretical work has shown that adverse selection does not necessarily lead to pecking order behavior. The pecking order is obtained only under special conditions. With both risky debt and equity being issued, there is often scope for many equilibria, and there is no clear basis for selecting among them. A pecking order may or may not emerge from the theory. Several articles show that the adverse selection problem can be solved by certain financing strategies or properly designed managerial contracts and can even disappear in dynamic models. Although adverse selection can generate a pecking order, it can also be caused by agency considerations, transaction costs, tax consideration, or behavioral decision-making considerations. Under standard tests in the literature, these alternative underlying motivations are commonly observationally equivalent.

### The Pecking Order Structure

Myers (1984) developed the pecking order theory of corporate capital structure to address the puzzles created by the well-known financing irrelevance proposition of Modigliani and Miller (1958). The pecking order has been highly influential, and it is an integral part of the literature that attempts to find a logically coherent and empirically successful theory of corporate financing.

Myers (1984) argued that issuing securities is subject to an adverse selection or “lemons” problem. Managers wish to use their private information to issue risky securities when they are overpriced. Hence, when a firm announces a new issue of risky securities, rational investors, who are at an information disadvantage, protect themselves by pricing these securities at a discount. Managers anticipate the discounting of the risky securities and so prefer to use securities that are immune to the lemons problem. As a result, the firm follows a pecking order: it finances with internal resources to the extent possible; if internal funds are inadequate, it obtains external debt, issuing risk-less debt followed by risky debt; firms issue equity only under duress or when needs exceed internal resources so that financing with debt would produce excessive leverage.

The development of the theory by Myers (1984) was motivated by Myers and Majluf (1984). The original presentation was primarily verbal, and it included several features that are not in the adverse selection model. As a result, the literature has adopted several distinct empirical interpretations of the theory, leading to alternative testing strategies.

#### Defining a Firm Flow Budget

Consider a firm that uses real fixed assets, working capital, and internal financial resources to produce revenue. It may be financed with debt, equity, or internal resources. Therefore, cash comes in from operations, external financing, and interest income on the firm’s financial investment. Cash goes out to invest in real fixed assets and working capital, pay interest on debt, and issue dividends. The firm’s cash holdings adjust so that the flow budget constraint is satisfied with cash-in equal to cash-out:

$Display mathematics$

In this expression, $Δxt$ denotes $xt−xt−1$ for any variable $xt$, where $t$ is the date. Let $ρt$ be the return on cash held by the firm and $rt$ be the return on debt owed by the firm. The budget constraint is an identity not a theory, and it holds because of the way these cash flow variables are defined.

Although this presentation is straightforward from a finance perspective, it does not match standard corporate accounting because both interest expense and interest income are part of operating cash flow in the firm’s financial statements. Therefore, internal cash flow is defined as:

$Display mathematics$(1)

Under this definition, there is a standard flow budget constraint:

$Display mathematics$(2)

Equation (2) is high dimensional in the sense that debt, equity, investment, working capital, dividends, and cash holdings are all commonly regarded as firm decisions.

Define the firm’s financing needs as:

$Display mathematics$(3)

The pecking order theory treats the elements of the needs as exogenous. The arrival of investment opportunities, which depends on business conditions, is the main determinant of investment. Working capital consists of inventory plus accounts receivable less accounts payable. These, too, are supposed to be driven by the real business side of the firm.

Dividends are assumed to be smoothed and primarily determined by past dividends. Evidence in support of dividend smoothing goes back to Lintner (1956). Whether it is sufficient to treat dividends as exogenous is debatable. Treating the change in cash holdings as exogenous is a heroic simplification. One could easily quibble with this treatment of cash because only a part of the firm’s cash may be required for business operations. Firms often hold cash for precautionary reasons, and they could finance some or all of the deficit by drawing down on their cash holdings. Firms without much cash holdings have little room to adjust without going to external financial markets.

Substituting Equation (3) into Equation (2), the budget constraint can be re-expressed as:

$Display mathematics$(4)

A financing deficit ($DEFt$) is defined as the difference between a firm’s needs ($Needst$) and its internal cash flow ($CFt$). This deficit must be financed externally through issuance of net debt ($ΔDt$) and net equity ($ptΔNt$).

$Display mathematics$(5)

The assumptions embedded in Equations (1) and (3) are common, but other ideas can change the definitions to some degree. As studies adopt slightly different definitions of $Needs$ and $CF$, they provide alternative interpretations of the verbal descriptions in Myers (1984). For example, one could argue about the distinction between long- and short-term debt, and where each belongs. Investment may depend on the cost of capital, which in turn may depend on the financing mix. One could also be concerned about how best to treat leases, pension obligations, swaps, options, and other derivatives issued by firms. These are important issues in their own right, and in some cases, they may break the pecking order. Rather than exploring a lengthy list of variations, a traditional and straightforward presentation of the pecking order is used.

#### What Are Real-World Flow Budgets Like?

Myers (1984) stressed the heavy reliance on internal finance and debt. He showed that the pecking order provided a plausible account of financing of firms between 1973 and 1982. How representative is that evidence? What is a “typical” or “representative” firm? Firms differ from each other in important respects; in particular, firm size heterogeneity is important. Large firms generally have easier access to corporate bond markets. In presenting averages, equally (value) weighting observations more closely reflects how small (large) firms finance. Which is representative? Both are. This distinction leads to different interpretations of how firms finance themselves, something long recognized in the capital structure literature as seen in Fama and French (2005, 2012) and Frank and Goyal (2003). These studies highlighted the changing role of equity finance, particularly after the 1980s. Accordingly, this article updates the evidence that motivated Myers (1984), but with greater emphasis on firm size for 1971–2018.

Table 1 presents summary statistics regarding Equation (1) for publicly traded U.S. firms in the S&P Compustat database for 1971–2018. The sample excludes regulated government services, financial firms, and firms with missing assets. The cash flow items are scaled by lagged assets and described in the Appendix. Column 1 presents value-weighted averages, whereas column 2 presents equally weighted averages.

#### Table 1. Cash Flow Budget Constraints, 1971–2018

 Firm Size Categories All All Small Medium Large VW EW Firms Firms Firms (1) (2) (3) (4) (5) $N$ 181,181 181,181 60,412 60,342 60,427 Cash Flow Budget (fraction of beginning assets) Investments $(Inv)$ 0.102 0.100 0.080 0.111 0.110 Change in working capital $(ΔWorkingCapital)$ 0.009 0.020 0.016 0.027 0.016 Cash dividends $(Div)$ 0.022 0.010 0.005 0.009 0.016 Change in cash $(ΔCash)$ 0.005 0.011 0.010 0.015 0.007 Needs $(Needs)$ 0.137 0.140 0.110 0.163 0.148 Internal cash flow $(CF)$ 0.122 0.065 −0.020 0.099 0.118 Debt issue $(DI)$ 0.084 0.098 0.076 0.105 0.113 Debt repayment $(DR)$ 0.064 0.080 0.062 0.089 0.088 Net debt issues $(ΔD)$ 0.020 0.018 0.014 0.016 0.025 Equity issue $(EI)$ 0.012 0.066 0.122 0.057 0.020 Equity repurchase $(ER)$ 0.016 0.010 0.005 0.010 0.014 Net equity issues $(pΔN)$ −0.004 0.057 0.117 0.048 0.005 Financing $(Financing)$ 0.137 0.140 0.110 0.163 0.148 Financing Decision of Firms (% of observations) Internal (both net issue < 5%) 78.2 68.4 64.2 69.4 71.7 Net debt issue (only net debt issue $≥5%$) 17.7 18.0 14.9 18.1 20.9 Net equity issue (only net equity issue $≥5%$) 2.6 10.4 16.2 10.0 5.0 Dual issue (both net issue $≥5%$) 1.5 3.2 4.7 2.6 2.3

Notes. The mean values of elements of cash flow budget constraint of publicly traded U.S. firms in Compustat are presentend for 1971–2018. The sample excludes regulated government services, and financial firms. All cash flow items are scaled by beginning-of-period assets. Column 1 reports value-weighted (VW) averages (where the observations are value weighted annually and then averaged across the years). Column 2 reports equally weighted (EW) averages. Columns 3–5 report the averages for three size-based portfolios defined by the highest, middle, and lowest tercile of the annual size distribution. Size is measured using real values of total assets. Firms are sorted into four categories based on major financing actions. These are: (a) “internal” if firms do not issue net debt or net equity in excess of 5% of assets in year $t−1$; (b) “net debt issue” if firms issue net debt in excess of 5% of assets in year $t−1$; (c) “net equity issue” if firms issue net equity in excess of 5% of assets in year $t−1$; and (d) “dual” if firms issue both net debt and net equity in excess of 5% of assets in year $t−1$. The bottom panel reports the percentage of firm-years that fall into each of the four groups for the overall sample and the three size-based portfolios.

A comparison of value-weighted and equally weighted averages of cash flow items provides a nuanced perspective on deficits and how firms are financed. Internal cash flows are 12.2% of lagged assets for value-weighted observations and exceed investment (10.2%) plus the change in working capital (0.9%). However, using equally weighted observations, internal cash flows are only 6.5% of lagged assets and exhibit a substantial shortfall in comparison to investment (10.0%) plus the change in working capital (2.0%). Consistent with these differences, the net equity issue is negative (−0.4% of lagged assets) for value weight and positive (5.7% of lagged assets) for equal weight, reflecting heavy use of equity financing by small firms. The net debt issues are comparable across the two weighting schemes. These differences between the two weighting schemes highlight the fact that small and large firms differ in the extent to which they rely on internal financing and debt to meet their investment needs. Large firms can generally finance investment and working capital internally, and they issue debt to meet any shortfalls. Small firms, in contrast, more often need external funds, primarily equity.

Next, $Needs$ is disaggregated into its components. For the average firm in the sample, investment in real assets is a critical driver of its funding needs. Regardless of the weighting scheme, the investment in fixed assets is about 10% of lagged assets, which is approximately 71% of the $Needs$. Other components of $Needs$ are relatively minor. Investments in working capital are roughly 1–2% of lagged assets. Cash dividends represent 2.2% for value weight and 1.0% for equally weighted firms, reflecting the differences in payout policies between large and small firms. Average $Needs$ as a fraction of lagged assets is about 14% in both cases.

Because of the importance of these size differences, firms are sorted annually into three size-based portfolios. Columns 3–5 of Table 1 show stark differences across the size categories. Small firms are unprofitable with negative internal cash flows (−2% of lagged assets), whereas large firms are highly profitable (11.8% of lagged assets). $Needs$ also differs significantly across size categories. Large firms have a substantially higher need for funds driven by large investments in both fixed assets and working capital. Hefty dividend payouts by large firms also contribute to their higher needs. Large firms primarily finance these needs internally and finance any deficit that remains with debt. They issue very little equity (0.5% of lagged assets). In short, when large firms access external capital markets, they primarily issue debt. This is pecking order behavior. Small firms, by contrast, primarily finance through equity. These net equity issues are significant and average about 12% of lagged assets. Small firms also issue debt but not as much as the equity that they issue.

The bottom four rows of Table 1 present the frequency of major financing actions. A 5% threshold is used to determine whether an intentional capital structure change has occurred (e.g., Hovakimian, Opler, & Titman, 2001). Firms are categorized into (a) those financing internally (both net debt and equity issue $<5%$ of lagged assets), (b) issuers of net debt (net debt issue $≥5%$ of lagged assets), (c) issuers of net equity (net equity issue $≥5%$ of lagged assets), and (d) dual issuers (both net debt and equity issue $≥5%$ of lagged assets).

Firms do not adjust financing on a large scale much of the time, but when they do, it is more often in debt markets than in equity markets. Major net debt issues are more common than major net equity issues for both value weight and equal weight. Across size categories, large firms rely on internal financing more often than small firms do. Small firms issue both equity and debt as they have substantial financing needs and negative internal cash flows. Large firms are relatively less likely to issue equity and more likely to issue debt.

In summary, large firms rely significantly on internal financing to meet their needs. External net debt issues finance the minor deficits that remain. Equity is not a significant source of financing for large firms. By contrast, small firms lack sufficient internal resources and finance externally. Although much of it is equity, there are substantial issues of debt by small firms.

#### Defining the Pecking Order

The firm’s flow budget constraint shows how the real and the financial fit together. That connection is definitional. The original pecking order theory relies on adverse selection, along with several exogeneity assumptions. The endogenous variables are debt $Dt$ and the number of shares $Nt$. The standard pecking order says that there are three distinct tiers in which the firm uses different ways to finance itself:

1.

Pecking Order Tier 1. Suppose that $Needst≤CFt$. Then $ΔDt=0$ and $ΔNt=0$. In this case, the firm has enough internal resources to fund its needs, and it must decide how to distribute its surpluses (i.e., $CF$ over $Needs$). It can pay down debt, or it can repurchase equity. Although the usual presentations of the pecking order theory often ignore the capital distribution problem, Myers (2003) pointed out that firms with surplus work up the pecking order when distributing money. The argument is that managers of overvalued firms refrain from repurchasing shares at too high a price. Therefore, if a firm does repurchase equity, investors assume that managers have positive information, not yet reflected in stock prices, causing prices to rise. Given the price impact of attempted stock repurchases, firms instead use their surplus to pay off debt. Firms repurchase equity only after the outstanding debt has been repaid. Although de Jong, Verbeek, and Verwijmeren (2010) showed that firms with surpluses use a large fraction of their surplus to repurchase debt, it is unclear whether they repurchase equity only after the debt has been paid off.

2.

Pecking Order Tier 2. Suppose that $Needst>CFt$ and $Needst≤CFt+DC¯t−Dt−1$. Then $ΔDt=Needst−CFt>0$, and $ptΔNt=0$. In this case, internal cash flows are not sufficient to meet the firm’s needs, and it must close the financing deficit through external funds. The pecking order says that debt is used if possible. Debt is adequate to close the gap as long as the debt capacity ($DC¯t$) does not bind. Debt can never exceed capacity, so $Dt≤DC¯t$, or $ΔDt≤DC¯t−Dt−1$. In this tier, the debt capacity does not bind and no equity is issued.

3.

Pecking Order Tier 3. Suppose that $Needst>CFt+DC¯t−Dt−1$. Then $Dt=DC¯t$ and $ptΔNt=Needst−CFt−DC¯t+Dt−1>0$. Because the debt capacity is binding, the firm sets $Dt=DC¯t$, and equity issues, $ptΔNt$, must make up the balance of the necessary funding. Thus, the final rung of the pecking order says, $ptΔNt=Needst−CFt−DC¯t+Dt−1$. The firm uses internal resources plus the full debt capacity plus some new external equity to fund itself.

The pecking order does not specify how many firms are in each tier. The theory pertains to individual firms and is silent about the aggregate. The literature often implicitly assumes that all or at least most firms are in Tier 2 (e.g., Frank & Goyal, 2003).

Over the last five decades, both the corporate sector as a whole and even most individual firms have not drifted to a corner with just a single type of financing. In the aggregate, there is a fair degree of long-run stability of capital structure across many decades and many countries as documented by Frank and Goyal (2008) and Graham, Leary, and Roberts (2015). At the firm level, however, there is much less stability as shown by DeAngelo and Roll (2015). These facts do not readily emerge from the traditional pecking order theory.

If a firm issues debt or sells new shares in the market, its leverage ratio changes. The market price of its shares reflects these security issuing decisions. If firms follow the pecking order, there may be new asset pricing implications from the nature of the time-varying leverage and investments. No study explores this issue carefully.

There are, however, short-run event studies that examine the market reaction to security issues. Debt issues generally have little effect on share prices, whereas common share issues are associated with share price drops (e.g., Eckbo, Masulis, & Norli, 2008). Although one could conclude that the fall in share price upon the announcement of a new issue is consistent with the pecking order theory, the interpretations are not straightforward. Depending on the issuing motivation and terms, the new equity issue could increase or reduce the value of the pre-existing shares. If share prices drop when equity is issued, the information that the market has about the firm may have changed, the terms of the issue may not have been fair, or the market may not be strictly efficient. Any connection to the pecking order is a loose one.

### Pecking Order Statistics

#### Financing Actions by Surplus and Deficit Firms

It is natural to ask whether the cash flow budgets of firms conform to the predictions of the tiers suggested by the pecking order theory. Firms are sorted into three portfolios based on whether they have a surplus or a deficit. Firms generate surpluses if $CF$ exceeds $Needs$ by more than 5% of lagged assets. Firms are in balance when $Needs−CF$ is within $±5%$ of lagged assets. Finally, firms generate deficits when $Needs$ exceeds $CF$ by more than 5% of lagged assets. Firms with surpluses and those in balance are in pecking order Tier 1, whereas firms with deficits are in pecking order Tiers 2 and 3. Table 2 presents the results.

#### Table 2. Cash Flow Budgets: Deficit Versus Surplus Firms

 Generating Generating Generating Surplus Neither Deficits (1) (2) (3) $N$ 26,797 101,708 52,676 Cash Flow Budget (fraction of beginning assets) Investments $(Inv)$ 0.031 0.067 0.198 Change in working capital $(ΔWorkingCapital)$ −0.019 0.010 0.058 Cash dividends $(Div)$ 0.008 0.012 0.008 Change in cash $(ΔCash)$ −0.008 −0.004 0.048 Needs $(Needs)$ 0.012 0.085 0.312 Internal cash flow $(CF)$ 0.116 0.088 −0.004 Debt issuance $(DI)$ 0.061 0.047 0.217 Debt repayment $(DR)$ 0.142 0.051 0.103 Net debt issues $(ΔD)$ −0.081 −0.005 0.113 Equity issuance $(EI)$ 0.009 0.008 0.208 Equity repurchase $(ER)$ 0.032 0.006 0.005 Net equity issues $(pΔN)$ −0.023 0.002 0.203 Financing $(Financing)$ 0.012 0.085 0.312 Financing Decision of Firms (% of observations) Internal (both net issue $<5%$) 95.8 95.4 2.4 Net debt issue (only net debt issue $≥5%$) 1.8 2.1 56.8 Net equity issue (only net equity issue $≥5%$) 2.4 2.5 29.8 Dual issue (both net issue $≥5%$) 0.0 0.0 11.1

Notes. Firms are sorted into three deficit groups: (a) firms with surplus $(Needs−CF<−0.05)$; (b) firms that are in balance $(−0.05≤Needs−CF≤0.05)$; and (c) firms with deficits $(Needs−CF>0.05)$. Within each deficit group, firms are further sorted into four categories based on major financing actions. These are: (a) “internal” if firms do not issue net debt or net equity in excess of 5% of assets in year $t−1$; (b) “net debt issue” if firms issue net debt in excess of 5% of assets in year $t−1$; (c) “net equity issue” if firms issue net equity in excess of 5% of assets in year $t−1$; and (d) “dual” if firms issue both net debt and net equity in excess of 5% of assets in year $t−1$. The bottom four rows report the percentage of firm-years that fall into each of these four categories for the three portfolios sorted on financing deficit.

About 15% of firm-year observations are in the surplus group. Surpluses arise for various reasons. Low investment needs (3.1% of lagged assets) and minimal dividend payouts (0.8% of lagged assets) reduce the needs. At the same time, these firms reduce working capital (−1.9% of lagged assets) and deplete cash holdings (−0.8% of lagged assets), leading to surpluses. These firms also exhibit high levels of internal cash flows (11.6% of lagged assets). In summary, low investment needs and high profitability coupled with reductions in working capital and depletion of cash holdings lead to large surpluses.

Firms use surpluses to both pay down debt and to repurchase equity. The net debt issues are −8.1% of lagged assets, and the net equity issues are −2.3% of lagged assets. The fact that firms primarily use surpluses to pay off debt may suggest that firms care about building debt capacity for future financing. The pecking order theory, however, makes a stronger prediction. According to Myers (2003), firms should repurchase equity only after they have paid down their debt. But in reality, firms often repurchase equity while they still have significant debt remaining on their balance sheet. Non-pecking-order motives for capital distribution policies appear to be at work.

About 56% of firm-year observations are in the balance group, as shown in column 2 of Table 2. These firms generate internal cash flows (8.8% of lagged assets), which are just about enough to meet their investment and dividend needs (8.5% of assets). Predictably, these firms are the least active in external financial markets. They issue just enough debt to meet their debt repayments. They are relatively inactive in equity markets, with both issuances and repurchases being close to zero. Firms in this group maintain net financing close to zero.

About 29% of firm-year observations are in the deficit group. These firms have significant investment needs (19.8% of lagged assets) and large working capital requirements (5.8% of lagged assets). Their deficits increase further because of a build-up of cash holdings (4.8% of lagged assets). Importantly, these firms are not profitable, and their internal cash flows are about zero (−0.4% of lagged assets). Thus, deficits arise because of a combination of negative profitability and significant investments in both real and financial assets. Firms finance these deficits in external capital markets: net debt issuances of 11.3% of lagged assets and net equity issuances of 20.3% of lagged assets.

Even firms generating deficits pay dividends, roughly comparable to those paid by surplus-generating firms. This is surprising given the significant need for funds for investment and the necessity of raising capital. Deficit firms repurchase little equity, so much of their equity issuances are to cover their deficits, including financing of dividend payouts. Farre-Mensa, Michaely, and Schmalz (2014) discussed reasons for financed payouts, including a desire to manage capital structure and cash holdings, monitor managers, and engage in market timing. Although the firms generating deficits raise funds more through debt than through equity, the net issues of debt are smaller than equity because firms must repay some debt because of its fixed maturity.

In summary, firms generating surpluses return money to both debt and equity holders. For self-sufficient firms, their issuance activities on the debt side are just large enough to offset their debt repayments. These firms are not active in equity markets. Firms generating deficits (because of lack of profitability and large investment needs) access both debt and equity markets. The average firm issues more equity than debt. However, a focus on major financing decisions reveals a higher propensity to make major net debt issues rather than major net equity issues.

#### Financing Actions by Deficit Size

According to the pecking order theory, firms generating deficits issue debt if they have not reached their debt capacity and issue equity if they have. However, it is not clear whether firms with deficits (as in column 3 of Table 2) have reached their debt capacity. The notion of debt capacity is theoretically ambiguous. If debt capacity is simply the optimal debt level from the firm’s perspective or an outcome of market equilibrium, it may not be a feasibility constraint at all. If debt capacity reflects the costs and benefits of optimal debt, the distinction between the pecking order theory and a conventional trade-off theory is ambiguous.

No entirely satisfactory resolution is available in the literature. Lemmon and Zender (2010), for example, used several conventional firm attributes to estimate a firm’s probability of having a debt rating and apply this probability as a proxy for debt capacity. Because the firm attributes that determine ratings are similar to those that affect leverage, it is possible that firms with sufficient debt capacity are also those operating below their target leverage. Then, how are pecking order predictions distinguished from predictions based on forces like debt overhang? Leary and Roberts (2010) used the data to determine whether they could infer debt capacity from the observed decisions in a manner that justifies the pecking order. No sharply defined debt capacity emerged.

Instead of relying on an ad hoc proxy, firms with positive deficits are sorted into three portfolios based on the size of the deficit. The results are reported in Table 3. Firms with high deficits are more likely to have used up their debt capacity to meet their financing needs, so they tend to be in Tier 3. The debt capacity for firms with low deficits is less likely to be binding, so these firms tend to be in Tier 2. In a related test, de Jong, Verbeek, and Verwijmeren (2010) also sorted firms into those with normal deficits and those with large financing deficits. They showed that small firms have more large deficits and fewer surpluses, which explains the empirical findings that small firms issue more equity than large firms do.

#### Table 3. How Important Is the Size of the Financing Deficit?

 Size of Deficit Categories Low Medium High Firms with Deficit Deficit Deficit Deficit (1) (2) (3) (4) $N$ 17,528 17,558 17,590 52,676 Cash Flow Budget (fraction of assets) Investments $(Inv)$ 0.117 0.170 0.309 0.198 Change in working capital $(ΔWorkingCapital)$ 0.029 0.045 0.100 0.058 Cash dividends $(Div)$ 0.009 0.008 0.005 0.008 Change in cash $(ΔCash)$ −0.003 0.004 0.144 0.048 Needs $(Needs)$ 0.152 0.227 0.558 0.312 Internal cash flow $(CF)$ 0.075 0.047 −0.132 −0.004 Debt issuance $(DI)$ 0.144 0.214 0.291 0.217 Debt repayment $(DR)$ 0.089 0.107 0.114 0.103 Net debt issues $(ΔD)$ 0.054 0.107 0.178 0.113 Equity issuance $(EI)$ 0.028 0.078 0.516 0.208 Equity repurchase $(ER)$ 0.005 0.005 0.005 0.005 Net equity issues $(pΔN)$ 0.023 0.073 0.512 0.203 Financing $(Financing)$ 0.152 0.227 0.558 0.312 Financing Decision of Firms (% of observations) Internal (both net issue $<5%$) 7.1 0.0 0.0 2.4 Net debt issue (only net debt issue $≥5%$) 73.1 65.6 31.8 56.8 Net equity issue (only net equity issue $≥5%$) 19.6 25.8 43.9 29.8 Dual issue (both net issue $≥5%$) 0.2 8.6 24.3 11.1

Notes. The firm-year observations in Table 2 identified as having a financing deficit are included here. They are sorted into three groups: (a) low deficit, (b) medium deficit, and (c) high deficit. The same statistics as in Table 2 are reported for each group.

Column 1 of Table 3 reports that firms with low deficits have $Needs$ of 15.2% of lagged assets. These are almost twice as large as internal cash flow. The resulting deficit of 7.7% of lagged assets is financed through both debt and equity, with net debt issues contributing 5.4% and equity contributing 2.3%. Firms in this tier primarily finance with debt (about 73% of these firms have net debt issues exceeding 5% of lagged assets). Many firms in the low-deficit group are also active in equity markets, with about 20% of them issuing net equity over 5% of lagged assets. Equity issues by these firms are surprising from the perspective of the pecking order theory if firms in the lowest tercile of deficit are those with ample debt capacity.

For firms with medium deficits, both debt and equity finance the deficit of 18% of lagged assets, with net debt issues contributing 10.7%, and net equity issues contributing 7.3%. All of the firms in the medium-deficit group make major interventions in capital markets, with 65.6% of firms completing a major debt issue, 25.8% completing a major equity issue, and 8.6% issuing both major net debt and net equity.

Firms with high deficits, by construction, require massive interventions in external capital markets. The large deficits are primarily due to large investments (31% of lagged assets), net working capital of (10% of lagged assets), and build-up of cash holdings (14.4% of lagged assets). This cash accumulation in the face of high real investment needs and negative profitability is striking. The critical question is: how are the large deficits funded? Unlike other groups, high-deficit firms rely much more on equity than they do on debt. The net debt issues are significant (17.8% of lagged assets), but the net equity issues are remarkably large (51.2% of lagged assets). In terms of major financing decisions, about 44% of firms in this group issue net equity over 5%. The patterns reported here are consistent with the evidence in Denis and McKeon (2018) and Huang and Ritter (2020) who showed that firms with persistent cash needs issue equity and save much of it in cash. Firms accumulate cash to prepare for future financing needs and relax future financing constraints. Furthermore, Denis and McKeon (2018) showed that firms financing their persistent losses with equity are high-growth firms with low leverage and high investment in intangibles.

The evidence that 32% of high-deficit firms issue primarily debt and 24% issue both debt and equity is surprising from the perspective of the pecking order theory. Eckbo and Kisser (2018) also found that firms often finance capital-intensive investment through debt issues. Within the pecking order, this should mean that even high-deficit firms have not reached their debt capacity, so they issue debt. Alternatively, it could mean that other considerations guide financing choices so that firms issue both debt and equity to stay in balance. Or perhaps they are concerned about other frictions.

The sorting into terciles does not provide enough granularity. Figure 1, therefore, plots results from sorting firms into 100 equally sized bins based on their financing deficits. It shows the average net debt issue and net equity issue for firms in each bin. The dashed vertical line is the point of balance where the deficits are zero. The bins to the left (right) are firms with surpluses (deficits). The upper graph presents the financing decisions of all firms in the sample. Although it is useful to examine firms at the extremes, it is also instructive to examine firms in the middle. The lower graph, therefore, focuses on firms in the middle between the 30th and 60th percentiles of financing deficits.

In the upper graph of Figure 1, note the asymmetry: the surpluses are never as large as the deficits. This may reflect the possibility that some of the items in $Needs$ are not entirely exogenous. Thus, surpluses never get very large because firms increase investments, dividends, or cash holdings as they experience high internal cash flows. When firms distribute these surpluses, distributions are to both debtholders and stockholders, with debt repayments dominating equity repurchases over the entire range of surpluses. Equity repurchases increase with surpluses, but their magnitudes are consistently smaller than debt repayments.

Turning to the deficits on the right of the dashed vertical line in Figure 1, both debt and equity issues increase with deficit until they reach levels that are in the 90th percentile. At that point, there is a sharp increase in equity issues and a steep decline in debt issues. Overall, the figure suggests that firms are relatively more active in debt markets as long as the deficits are reasonable. They turn to equity markets when deficits become massive.

The lower graph of Figure 1 shows an apparent lack of discontinuity in financing choices from surplus firms on the left of the dashed vertical line to those with deficits on the right. Firms with small surpluses often issue small amounts of equity. Surprisingly, even firms with small deficits issue some equity. As deficits increase, net debt issues eventually become larger. But this sequence of financing does not seem to reflect pecking order considerations. The fact that firms do issue or repay equity in small amounts is emphasized by Fama and French (2005) as being problematic for the pecking order. It also raises questions about the commonly used 5% threshold as a measure of material external financing.

#### Which Firms Do Nothing?

Some firms do nothing to adjust leverage. To understand how these firms fit into the theory, start with firms that have a deficit of zero or $ΔDt+ptΔNt=0$ as in Equation (5). Without further theoretical assumptions, the adding-up constraint places no limitations on either net debt issues or net equity issues. According to the pecking order, firms should take no financing actions when deficits are zero, so $ΔDt=ptΔNt=0$. Furthermore, small surpluses should lead to a repurchase of debt, so $ΔDt<0$ and $ptΔNt=0$, whereas small deficits should lead to an issue of debt, so $ΔDt>0$ and $ptΔNt=0$. Therefore, there should be a point of discontinuity for debt issuance when the deficit is zero. Equity change should be zero for an extensive range that starts where the debt is entirely paid off (surplus) and continues to the point where the debt capacity is reached (deficit).

Figure 2 shows the financing actions around the zero deficit point. As in Figure 1, firms are sorted into 100 bins, and for each bin, Figure 2 shows: (a) the fraction of firms with precisely zero change in the debt over the year and (b) the fraction of firms that have precisely zero change in the number of shares outstanding over the year.

The zone of elevated inaction exists in a narrow band between bins 43 and 53. That is roughly from a financing surplus of around 2% to a deficit of 0.03%. At the point of zero deficit, about 50% of the firm-year observations have zero change in debt, and about 60% have zero change in the number of shares outstanding. This finding does not support the pecking order theory. The debt capacity would have to be reached at group 53, which is below the range at which most debt is issued. Figure 1 suggests that the debt capacity is at roughly bin 90.

Roughly half of the firm-year observations have non-zero financing actions when the deficit is zero. Because $ΔDt+ptΔNt=0$, this means that firms take offsetting financing actions. For firms with small surpluses (deficits), about 80% (90%) of them have changes in the number of shares outstanding. Again, this does not accord well with the pecking order prediction.

Note that, in trade-off models (e.g., Fischer, Heinkel, & Zechner, 1989; Frank & Goyal, 2015; Goldstein, Ju, & Leland, 2001), fixed cost of leverage adjustment implies that there exists a zone over which firms make no active leverage adjustments. Leverage simply drifts as a reflection of the changes in the market value of equity. If firms in the inaction zone in Figure 2 also happen to be away from their optimal leverage, this feature of the data is consistent with the fixed-cost adjustment in the trade-off models. But again, the zone of inaction seems narrow.

#### Which Firms Undertake Major Financing Actions?

How much do financing deficits drive the observed financing actions of firms? The distribution of firms across deficit categories is examined for firms financing internally and firms financing externally in the form of major debt issues, major equity issues, and dual issues (firms issuing both debt and equity). Table 4 reports that of the firms financing internally, most are in balance (78%) or generating a surplus (21%). The pecking order does not predict that these firms should issue either debt or equity because the internal cash flows are either about equal to their needs or exceed them. Among the firms financing internally, few are generating a deficit but this deficit is smaller than the threshold used to track debt and equity issues.

#### Table 4. Distribution Across Deficit Categories for Firms Choosing Particular Financing Decisions.

 Deficit Categories Intensity of Deficit Generating Generating Generating Low Medium High Surplus Neither Deficit Deficits Deficits Deficits (1) (2) (3) (4) (5) (6) Internal (N = 123,973) 20.7% 78.3% 1.0% 1.0% 0.0% 0.0% Debt (N = 32,536) 1.5% 6.5% 92.0% 39.4% 35.4% 17.2% Equity (N = 18,851) 3.4% 13.4% 83.2% 18.3% 24.0% 41.0% Dual (N = 5,821) 0.0% 0.0% 100.0% 0.6% 25.9% 73.4

Notes. Four subsamples of firms are examined. “Internal” examines firms that issued neither net debt nor net equity in excess of 5% of beginning assets. “Debt” examines firms that issued net debt in excess of 5% of beginning assets. “Equity” examines firms that issued equity in excess of 5% of beginning assets. “Dual” examines firms that issued both net debt and net equity in excess of 5% of beginning assets. Columns 1–3 report the percentage of firms in each subsample that generate a surplus, generate neither surplus or deficit, or generate a deficit. Columns 4–6 further classify deficit firms into three groups based on the intensity of their deficit. Low-, medium-, and high-deficit firms are the bottom, middle, and top tercile of deficit firms, respectively.

Debt issuers are typically firms with deficits. However, most of these are by firms that have low or medium deficits. Firms in the highest tercile of deficits more often fund their deficits with equity than with debt. Dual issuers are all concentrated among deficit-generating firms. Dual issues become more common among higher deficit terciles.

In summary, firms generating surpluses generally do limited external fundraising. Firms roughly balanced between cash needs and internal cash flow commonly undertake small amounts of debt and equity issuing. They more often issue equity than debt. Firms with deficits are active in both debt and equity markets. As deficits increase, the likelihood of major net debt issues falls as the likelihood of major net equity issues rises.

### Theory

The original justification for the pecking order is adverse selection. Myers and Majluf (1984) developed the adverse selection theory of financing, and Myers (1984) combined it with additional ideas in the pecking order. The adverse selection models following Cadsby, Frank, and Maksimovic (1990) are first discussed. Bolton and Dewatripont (2005) and Tirole (2010) provided good textbook versions of the theory. This section also discusses how pecking order financing can arise even in the absence of adverse selection.

Some financial economists choose to interpret the pecking order theories literally. They point to evidence that is either consistent or inconsistent with a range of implications of the theory. Other scholars think of the pecking order as an interesting parable that can provide insight. Such scholars are generally much less interested in tests of the literal empirical implications of the theory. Which features of the theory appear more relevant often depend on tastes.

A firm is run by managers who know the true value of the firm’s assets and a profitable new project. Outside investors can only guess.

The firm is equity financed and has assets in place $Ai$ and a positive net present value project that pays off $Bi$ where $i$ is either high $(H)$ or low $(L)$. Therefore, $AH+BH>AL+BL$, and nature determines that the two types are equally likely. To undertake the project requires an equity investment of $I$ that must come from outside investors, or the opportunity is lost.

If the project happens, the value of the firm is $Vi=Ai+Bi+I$. But it must be shared between the original owners who get $(1−s)Vi$ and the new equity investors who get $sVi$. Does it pay for the manager to undertake the project or let it go? That depends on $s$. The value of $s$ depends on what the investors believe, as they must expect to at least break even. If $I, the investors gain, and if $I>sVi$, the investors lose money. Investor competition implies $I=E(sVi)$, where $E(⋅)$ is the expectation operator.

In general, there can be a pooling equilibrium, a separating equilibrium, or a semi-separating equilibrium. In a pooling equilibrium, both types of firms undertake the project. In this case, the investor knows that there is a 50% chance that the firm is type $L$ and a 50% chance that the firm is type $H$.

In a separating equilibrium, only the type $L$ firm undertakes the project. Accordingly, the investor knows that the firm raising the equity is type $L$ and prices the investment, $s∗=I/VL$. Investors treat any firm issuing equity as type $L$. The type $H$ firm prefers to forgo the new project because it does not get enough from the project to overcome the cost of losing a significant fraction of the current value by being treated as type $L$.

Can this separating equilibrium happen? The answer provided by Myers and Majluf (1984) is yes for some parameter values, and it has important implications for the pecking order. Self-financing would be attractive because it completely avoids the guessing game and potential adverse selection problem. This is the first tier in the pecking order hierarchy. If the debt is risk free, there is no adverse selection, and it is just as good as internal funding. Many articles assume that firms only issue safe debt, or that it is limited by the value of the firm’s collateral as in Hennessy and Whited (2005). These are convenient simplifications for theoretical analysis. But empirically, many firms do, of course, issue risky debt. The adverse selection cost for risky debt is not as large as that for equity; hence, firms choose risky debt over equity. Myers (1984) suggested that risky debt is somewhere between pure self-financing and equity financing and constitutes the second tier of the pecking order hierarchy.

Risky debt is conceptually complex. Although Myers and Majluf (1984) considered financing with risky debt, neither they nor Myers (1984) clearly defined debt capacity for risky debt. Lemmon and Zender (2019) extended the model of Myers and Majluf (1984) by considering debt covenants. Debt covenants allow the lender to liquidate the firm in certain scenarios, but liquidation is costly because of the lender’s inferior information. The liquidation value becomes a measure of the firm’s debt capacity in the sense that at that point, the entrepreneur has no motivation to issue debt in response to the adverse selection problem. The optimal level of debt in the model may be below the firm’s debt capacity, depending on assumptions about the renegotiation of covenants.

The natural debt limit from macroeconomics (e.g., Ljungqvist & Sargent, 2018) could be adopted and is determined by the maximum amount that a firm can repay if it uses all future cash flows to make payments on debt. This debt capacity is a feasibility constraint. But incentive issues probably matter. Tighter limits may reflect sequential incentives to service the debt. Whether such sequential incentives help produce pecking order behavior is likely to depend on detailed assumptions. Outside equity is not an equilibrium security. It is issued once the firm has reached its debt capacity. This defines the third tier of the pecking order hierarchy.

It is essential to recognize that even the retained earnings part of the pecking order may not be as simple as it sounds. The claim that self-financing completely avoids the information problems may be wrong. Tirole (2010) observed that the manager still needs to convince investors not to demand their money back. Although such a demand may sound unusual, it can happen directly through investor activism. For example, in the 1990s, Kirk Kerkorian, Chrysler’s largest shareholder, famously demanded that the company disgorge its substantial cash holdings to its investors (Marcum, Martin, & Strickland, 2011). Dealing with such a demand can raise similar adverse selection problems as raising external equity. In both cases, management must convince an outside investor about claims on the firm, and management presumably has better information. Therefore, retained earnings may not always be so free of informational issues.

With both risky debt and equity being issued, there is often scope for many equilibria, and there is no clear basis for selecting among them. Pecking order may or may not emerge from the theory—details matter, including equilibrium selection. The standard analysis of this issue was provided by Noe (1988). Cadsby, Frank, and Maksimovic (1998) provided experimental tests of the underlying theory, as well as tests of equilibrium selection. They found that learning and path dependence seem important. The standard equilibrium selection criteria seem less helpful.

Several articles showed that adverse selection problem could be solved by certain financing strategies (Brennan & Kraus, 1987; Chakraborty & Yilmaz, 2011; Constantinides & Grundy, 1989) or properly designed managerial contracts (Baranchuk, Dybvig, & Yang, 2010; Dybvig & Zender, 1991) and could even disappear in dynamic models (Morellec & Schürhoff, 2011).

The theoretical work has shown that adverse selection does not necessarily lead to pecking order behavior. The pecking order is obtained only under special conditions, as in Nachman and Noe (1994).

An important point is whether the information asymmetry is about assets in place or growth opportunities. The pecking order, as in Myers and Majluf (1984), arises when the information asymmetry is about assets in place. Both assets in place and growth opportunities are distinct, and each could have a high degree of private information. Myers (2015) emphasizes that pecking order should apply to mature firms and not to growth firms because “it’s mostly managers’ inside information about assets in place that blocks equity issues and investment” (p. 13). Although it is true that mature firms often have relatively more assets in place than growth opportunities, it is unclear whether information asymmetries overall are more significant for mature firms. More information has accumulated about assets in place for mature firms. Small, young firms are more likely to have greater information asymmetries about all of their assets because of their short history.

A more subtle perspective on how asymmetric information affects the use of debt and equity is Fulghieri, Garcia, and Hackbarth (2020). Here, the relative mispricing of debt and equity depends on the comparative exposure to the firm’s assets in place, growth options, and volatilities. The emphasis on relative volatility is an interesting contribution of this article. Growth options are risky, but perhaps less exposed to asymmetric information than are assets in place. As a result, there can even be a reversal of the traditional pecking order. In their model, violation of the usual pecking order is particularly likely for small firms, which is observed in the data. This article contributes to the growing recognition that adverse selection does not automatically imply the usual pecking order.

Halov and Heider (2011) pointed out that because debt is a concave claim, investors who are uninformed about risk could misprice it. Because the firm must issue those securities that are likely to be least mispriced by imperfectly informed investors, they will sometimes avoid issuing debt. This could happen if the odds of default are significant, and debt issues can create information problems. The adverse selection cost matters most for small, young, and non-dividend-paying firms as the outside market knows less about their risk. Hence, it is not surprising that these firms issue equity even though they have not exhausted their debt capacity.

There is an interest in how ambiguity aversion may affect corporate finance (see Izhakian, Yermack, & Zender, 2016; Malenko & Tsoy, 2019). The basic idea is that investors believe that many things can happen, and they focus on the worst path when valuing assets. If outsiders think that the worst path is worse than what the insiders know, they will undervalue that security relative to what insiders know even if everyone has the same beliefs about the expected values. Under ambiguity aversion, large uncertainty can be different from small uncertainty. Malenko and Tsoy (2019) found that in the presence of ambiguity aversion and private information about a new project, for sufficiently large uncertainty, equity financing is used. For sufficiently low uncertainty, there is risky debt but no equity. Much like the standard adverse selection pecking order, if there is private information about assets in place, equity is avoided, and firms generally use debt financing.

Not all discrete financing region theories are pecking order theories because they may not have the standard tier structure. For example, Frank and Sanati (2019) studied a model based on tax and collateral considerations. In that model, firms initially issue equity for investment purposes, and later, they issue debt and repurchase the equity once the investment is in operation. That generates discrete financing behaviors, but it need not match the traditional pecking order predictions regarding the three tiers.

Bharath, Pasquariello, and Wu (2008) evaluated the core assumption of information asymmetry for the pecking order theory. They constructed several market microstructure asymmetry proxies and tested whether firms with more asymmetry use more debt finance. The asymmetry is time varying so the pecking order may be more applicable at some times than at other times. If this is right, there may be significant interactions with the market-timing perspective (e.g., Baker & Wurgler, 2002), which have not been studied.

#### Other Justifications

##### Taxes

The first distortion introduced into the Modigliani and Miller theorem is the tax deductibility of interest payments in Modigliani and Miller (1963). They do not distinguish between internal and external financing. All financing is external. They concluded that firms should be 100% debt financed to exploit the tax advantage fully. However, if a “debt capacity” limit somewhere below 100% is introduced, the firm may first use that capacity to issue debt. But if that does not raise enough money, the firm would turn to external equity. A pecking order may quickly emerge in such a model.

Stiglitz (1973) highlighted the internal versus external financing distinction. There is a fundamental asymmetry in the tax code. If an investor gives money to a firm, the investor does not pay a tax. If a firm returns money to an investor, the investor does pay a tax. The amount of tax paid depends on whether the investor is providing debt or equity to the firm. As a debt holder, the investor pays taxes on interest income, historically more heavily taxed than capital gains from equity. As a shareholder, the investor pays taxes on dividends and capital gains. Once funds are inside the firm, there is a tax incentive to leave them inside the firm even though firm profits are taxable at the corporate rate. Stiglitz (1973) observed that if the personal tax rate exceeds the corporate rate, it pays to finance as much investment as possible through retained earnings. Whether excess investment beyond retained earnings is funded by equity or debt depends on a firm’s after-tax return to investors. The observed leverage ratio is a “fortuitous outcome of the profit and investment history of the firm” (Stiglitz, 1973, p. 32). In other words, tax considerations may generate some firm decisions that are similar to the usual pecking order.

##### Agency

The claim that managers prefer internal financing to external financing is much older than the pecking order. Butters (1949) suggested that many executives naturally think of internal funds as low cost. A variety of factors appear to be responsible, including what are called agency considerations and behavioral factors.

Myers (2003) pointed out that the agency conflicts in Jensen and Meckling (1976) can generate a pecking order. An entrepreneur who self-finances the firm’s investment has no distortions. Both the marginal cost and the marginal benefit of the investment have their full impact on the entrepreneur’s wealth. Therefore, self-financing is undistorted. If debt is risk free, the investor still obtains the full marginal benefit of investing. Therefore, outside debt is just as good as self-financing. But outside equity is different. The entrepreneur pays the full cost of effort but shares the benefit of the effort with the outside investor. In this sense, equity is less desirable.

This underinvestment is inefficient. The use of internal financing results in higher welfare. Thus, retained earnings are preferred. Debt is just as good in this simple model. Equity is inefficient. Hence, it generates a version of the pecking order. If debt is relatively safe but not entirely risk free, it is easy to imagine that an ordinary pecking order with its three distinct categories could emerge.

Agency conflicts can give rise to pecking order behavior through other channels. For example, managers have incentives to meet earnings per share (EPS) targets because of their compensation schemes. When expected returns on investments are low, managers are reluctant to issue equity, which can dilute EPS. In this case, debt financing is preferred. Brennan and Kraft (2018) found empirically consistent evidence that managers rely more on debt financing when earnings prospects are poor.

Lambrecht and Myers (2012) studied a dynamic agency model in which managers attempting to maximize the rents they take from the firm make payout, investment, and financing decisions. Their model rationalizes the target payout model in Lintner (1956). In particular, they focus on the capital budget constraint, $ΔDebt+NetIncome=CAPEX+Payout$. Investment opportunities determine $CAPEX$, and the payout is smooth, so the change in debt, $ΔDebt$, must soak up most of the shocks in net income. Because debt absorbs fluctuations in operating profitability and capital investment, it follows a pecking order. Note that the aim of the agency issue is not to generate a pecking order of debt. It is the consequence of payout smoothing due to rent smoothing.

##### Behavioral

Heaton (2002) studied a model of corporate finance with overly optimistic managers and efficient capital markets. Optimistic managers believe that the capital market undervalues the firm, and hence they view external financing to be unduly costly. Therefore, optimistic managers use internal cash and risk-free debt first and prefer risky debt to all equity financing. It is challenging, however, to measure managerial optimism. Malmendier and Tate (2005) and Malmendier, Tate, and Yan (2011) constructed measures of chief executive officer (CEO) overconfidence. They argued that an overconfident CEO regards outside funds as too expensive, and this mainly affects external equity. Therefore, a pecking order may quickly emerge.

Alternatively, perhaps the pecking order is a simple heuristic that economizes on decision making. Malmendier et al. (2011) may be understood as reflecting heuristic decision making where the heuristics reflect CEO experiences. A CEO who adopts a suboptimal approach on her account often seems to take a similar (suboptimal) approach for firm decisions. Gigerenzer and Gaissmaier (2011) provided a helpful review of the literature about the role of heuristics.

Other behavioral biases can also generate non-pecking-order decisions. Hackbarth (2008) compared managers with biased growth perceptions to managers with biased risk perceptions. In one case, a pecking order emerges, and in the other case, a reverse pecking order is predicted. Therefore, the exact nature of the bias is essential.

##### Other Mechanisms

Studies including Altinkilic and Hansen (2000), Leary and Roberts (2010), and Fama and French (2012) suggest that transaction costs can also generate pecking order behavior. For example, pecking order could be a result of issuing costs that are zero for retained earnings, low for short-term debt, and highest for share issues.

Anderson and Carverhill (2011) studied a model of dynamic cash holdings and showed that the optimal cash policy leads to a financing hierarchy that depends on the business conditions. The standard pecking order holds if the business conditions are right. When the business conditions are adverse, firms prefer equity issues to short-term debt if internal funds are not available. This finding suggests that, in a dynamic pecking order framework, firms make debt choices after considering the costs of issuing securities over time. Firms expecting significant investments in the future preserve debt capacity.

### Key Tests

#### Shyam-Sunder and Myers Test

An influential approach to testing the pecking order is developed by Shyam-Sunder and Myers (1999) and known as the SSM test. Shyam-Sunder and Myers studied 157 firms that had continuous data from 1971 to 1989. The SSM test is a regression:

$Display mathematics$(6)

The key prediction is $βPO=1$, so Shyam-Sunder and Myers (1999) essentially assumed that all firms are in Tier 2. They estimated $βPO=0.85$, which is reasonably close to 1, albeit statistically significantly lower. This striking empirical evidence has generated considerable interest and effort to understand how best to interpret the results.

In a broader population of firms with a more extended sample period, Frank and Goyal (2003) found that $βPO=0.28$ and that empirically large firms behave more like the pecking order. This suggests that the sample for the SSM test may not be a good reflection of the broad population of firms. Their finding also raised the issue of the interpretation of the firm size effect. It is not clear whether the pecking order theory applies more strongly to small or large firms. Smaller firms typically receive much less news media attention and analyst coverage, so insiders often have an informational advantage when compared to otherwise similar large firms. Larger firms are commonly thought to be slower growing and may have more of their value in the form of assets-in-place rather than growth options. Asset-in-place may be more subject to adverse selection as compared to growth options. Hence, in principle, the firm size implications can go either way, depending on what control variables are included and which force is more important. Frank and Goyal (2003) and Fama and French (2005) argued that the asymmetric information problems are more serious for small firms than for large firms, so the evidence that large firms behave more like the pecking order conflicts with the theory. Myers (2015) sharply criticized this idea. A helpful clarification of these issues was provided by Fulghieri et al. (2020).

The theory section suggests that a pecking order could be due to adverse selection, agency conflicts, and perhaps other forces. The SSM test does not identify an underlying theoretical justification for a pecking order. Even if an SSM test were to substantiate a pecking order, it would still be silent about the underlying driving force.

Does the SSM test correctly identify parameters of interest within the pecking order theory? Not necessarily. Chirinko and Singha (2000) provided examples that call into question the interpretation of the SSM test. In one case, the pecking order is true, but the SSM test rejects it. In another, the pecking order is false, but the SSM test does not reject it. This is disconcerting.

Some of the problems with the SSM test are easy to see. Suppose that the pecking order is true. Even then, the theory does not indicate how to define debt capacity. Therefore, it is not clear to which tier any firm belongs. In Tiers 1 and 3, the theory predicts $bPO=0$. In Tier 2, the pecking order predicts $bPO=1$. Outside observers do not know how many firms in the data belong to each tier. A population regression yields a weighted average of the firms with a coefficient of $0$ and a coefficient of $1$. The estimated coefficient says something about how many firms are in Tier 2 relative to the overall population of firms. It does not say anything clear-cut about the merits of the pecking order relative to other theories. Estimates of $bPO$ are probably best regarded as simple descriptive statistics rather than as structural parameters that test any particular theory.

The SSM test has been an influential testing method in the literature. However, drawbacks to the method have been convincingly established. Accordingly, without modifications, this is probably no longer a suitable method for testing the pecking order against alternative theories of capital structure.

#### Fama and French Test

Fama and French (2002) linked dividend payout and leverage with various firm characteristics through three sets of regression models: (a) a target dividend payout model; (b) a dividend partial adjustment model; and (c) a leverage partial adjustment model. They investigated whether the coefficients of various firm characteristics are consistent with the trade-off and pecking order predictions.

Fama and French (2002) argued that more profitable firms and firms with fewer investments should have higher dividend payouts under the pecking order theory, and the evidence supports this prediction. They also found positive relations between leverage and firm size, and between dividend payout and size. They argued that these results are consistent with the pecking order theory.

According to the pecking order, more profitable firms have more internal resources, and these internal resources pay for investments and then pay off debt. As a result, firms have less leverage. Fama and French (2002) provided support for this idea, and several studies found similar evidence (see Bradley, Jarrell, & Kim, 1984; Eckbo & Kisser, 2019; Frank & Goyal, 2009; Harris & Raviv, 1991; Rajan & Zingales, 1995). The negative correlation between profitability and leverage is the single most cited fact in support of the pecking order theory. Frank and Goyal (2015) showed that the negative relation is a consequence of adjustment costs. Firms respond to profitability shocks by adjusting their debt; however, the response is not large enough to offset the mechanical change in equity. Hence, leverage ratios decline even though higher profitability results in a response in debt markets consistent with alternative theories.

According to the pecking order, firms do not have leverage targets, so the estimated speed of adjustment should be indistinguishable from zero. Fama and French (2002) found statistically reliable evidence that the speed is positive, but its magnitude is suspiciously small. Flannery and Rangan (2006) and Faulkender, Flannery, Hankins, and Smith (2012) found similar evidence. However, the structural interpretation of adjustment speed is subject to the debate (see Chang & Dasgupta, 2009; Elsas & Florysiak, 2015; Huang & Ritter, 2009; Hovakimian & Li, 2010; Leary & Roberts, 2005). Furthermore, Frank and Shen (2019) showed that mismeasured targets bias the estimated speed coefficient toward zero. When firm-specific time-series models are estimated, much faster speeds of adjustment are observed. Consistently, DeAngelo and Roll (2015) found that time-varying leverage targets best explain the observed leverage dynamics.

There is significant equity issuing by small low-leverage growth firms (Fama & French, 2002). The evidence suggests that small public firms raise significant amounts of equity in private markets through private investment in public equity (Gomes & Phillips, 2012; Phillips & Sertsios, 2017). Gomes and Phillips (2012) showed that small firms with high asymmetric information are particularly frequent issuers in private equity markets. Firms issue informationally sensitive securities to private investors with whom they can share non-public information more readily.

Overall, it seems clear that there is much more common use of equity financing since the 1980s in contrast to the data that motivated Myers (1984). To see whether this contradicts the pecking order requires identifying to which of the three tiers each firm-year belongs. This idea leads naturally to the tests by Leary and Roberts (2010).

#### Leary and Roberts Test

The SSM test does not adequately reflect the hierarchical structure of the theory. Leary and Roberts (2010) explicitly adjusted for the hierarchical structure of the pecking order. Because the pecking order does not precisely define the locations of the thresholds, they tried a variety of empirical proxies for the threshold locations. They also consider both a strict interpretation of the theory and more liberal interpretations in which various adjustments take into account other considerations.

Leary and Roberts (2010) reported that $67%$ of financing uses internal funds, $23%$ uses debt, and the rest uses equity. The dominance of internal funding is as expected in a pecking order. They also found that equity finance is particularly important for smaller and younger firms.

In a more liberal interpretation of the pecking order theory, Leary and Roberts (2010) found that the pecking order correctly predicts $74%$ of the internal–external financing splits, $30%$ of the debt–equity splits, and $0%$, that is, none, of the equity issuance decisions. Almost $39%$ of the debt issues violate the pecking order because they are undertaken even when internal funds exceed investment.

An innovation of Leary and Roberts (2010) is that they exploited the data directly to infer a set of implied thresholds for internal funding and for inferring debt capacity. Their results suggested that 60% of the equity issues take place when firms still have adequate debt capacity. For this to be reasonable, firms would have to be remarkably debt averse beyond what can be justified by considerations of the remaining investment grade.

Leary and Roberts (2010) examined which friction(s) generate(s) pecking order behavior and employ(s) several proxies for information asymmetry. However, they found no robust evidence that firms of high information asymmetry adhere more to the financing hierarchy. Instead, when they stratified the sample according to agency cost proxies, they found systematic and robust patterns indicating that firms with significant agency problems are more likely to adhere to the pecking order. They used firm characteristics, such as firm size, cash flow, and market-to-book ratio, as proxies for agency cost. Still, the economic interpretation of these variables is not clear-cut.

In summary, Leary and Roberts (2010) showed that the strict pecking order makes a particularly sharp prediction about equity issuance relative to other forms of financing. That prediction fails. Even adding further factors to generate an “extended pecking order” does not resolve the problem. The pecking order offers only minor improvements over a completely naive estimator based on no theory. Simple attempts to patch up the pecking order theory do not resolve the empirical problems, and they lose the beautiful simplicity of the pecking order.

### Where Do We Stand?

The pecking order theory was originally motivated by the idea that equity has a more serious adverse selection problem than debt. However, a pecking order structure can also emerge from other factors such as tax considerations, transaction costs, agency frictions, or behavioral factors. Furthermore, if adverse selection affects the second moment instead of the first, the ordering between debt and equity can be reversed. If the firm chooses a variety of financing contracts, there are commonly multiple equilibria. There is no assurance that a pecking order would emerge even if adverse selection affects equity. Therefore, evidence for or against the pecking order is not the same as evidence for or against adverse selection in financial markets.

How does the pecking order do empirically? Is it rejected, or not? The pecking order theory is, of course, rejected by some aspects of the data. But that is probably not the most critical question to ask about the pecking order. Strong assumptions are needed to generate the pecking order, and it provides a rather stark set of predictions. Stark predictions are interesting. They put ideas at risk. But their very starkness means that they likely will not reflect everything that is going on. Accordingly, the following questions seem more useful than a simple, reject or not, question:

1.

What features of the data make sense from a pecking order perspective?

2.

What features of the data are surprising from a pecking order perspective?

3.

To what extent does adverse selection seem to be the key driving force?

An answer to question 1 is that the pecking order provides a framework that helps explain several significant patterns in the data. Firms with moderate deficits favor debt issues. Firms with very high deficits rely much more on equity than on debt. These financing patterns make sense within the pecking order.

An answer to question 2 is that the pecking order has trouble with equity issues and with defining the boundaries at which financing switches. As documented in many articles, equity issues have become more prevalent since 1980s. Equity-repurchasing firms generally do not pay off all debt first. Many equity-issuing firms do not seem to have entirely used up “debt capacity” under any obvious definition of that capacity. There are even some equity issues by firms with an operating surplus. The pecking order suggests a sharp discontinuity of financing methods between surplus firms and deficit firms, and another sharp discontinuity at the debt capacity. The literature provides little or no evidence to support these threshold effects. More generally, the pecking order has trouble explaining the inaction firms depicted in Figure 2.

An answer to question 3 is challenging. There is no one-to-one connection between adverse selection and the pecking order. As a matter of theory, if the firm can issue both equity and risky debt, there are commonly multiple equilibria. Many of those equilibria do not exhibit pecking orders. Adverse selection can generate a pecking order. But it can also be caused by agency considerations, transaction costs, tax consideration, or behavioral decision-making considerations. Under standard tests in the literature, these alternative underlying motivations are observationally equivalent. Other kinds of tests are needed to distinguish among them. As a result, any claimed answer to question 3 is more a matter of taste and belief than a reflection of established evidence.

Starting with Myers (1984), the pecking order is often presented as a horse race with the trade-off theory as in Fama and French (2002) and Frank and Goyal (2008). That is an interesting contest. But a range of relatively subtle considerations is involved in making that comparison properly. The issues go deeply into the substance of the trade-off theory and thus are outside the scope of this article. The relative merits of the two approaches are not yet fully settled, decades after Myers (1984) started the race.

The pecking order provides an influential perspective on how firms may use external debt and equity finance. Some facts roughly align with the theory. However, other facts do not fit readily in the pecking order. The more literally the theory is interpreted, the greater the empirical challenges it seems to face. This problem may not be unique to the pecking order. A yet unmet challenge for any capital structure theory is to provide a fully satisfactory account of the key features of the data.

### Acknowledgments

The authors thank Michael Brennan, Robert Chirinko, Harry DeAngelo, Philip Dybvig, Espen Eckbo, Mark Flannery, Paolo Fulghieri, Thomas Noe, Gordon Phillips, Jay Ritter, Patrick Verwijmeren, and Jaime Zender for their helpful comments. All errors and misinterpretations are the responsibility of the authors.

#### References

• Altinkilic, O., & Hansen, R. S. (2000). Are there economies of scale in underwriting fees? Evidence of rising external financing costs. Review of Financial Studies, 13(1), 191–218.
• Anderson, R. W., & Carverhill, A. (2011). Corporate liquidity and capital structure. Review of Financial Studies, 25(3), 797–837.
• Baker, M., & Wurgler, J. (2002). Market timing and capital structure. Journal of Finance, 57(1), 1–32.
• Baranchuk, N., Dybvig, P. H., & Yang, J. (2010). Renegotiation-proof contracting, disclosure, and incentives for efficient investment. Journal of Economic Theory, 145(5), 1805–1836.
• Bharath, S. T., Pasquariello, P., & Wu, G. (2008). Does asymmetric information drive capital structure decisions? Review of Financial Studies, 22(8), 3211–3243.
• Bolton, P., & Dewatripont, M. (2005). Contract theory. Cambridge, MA: MIT Press.
• Bradley, M., Jarrell, G., & Kim, E. (1984). On the existence of an optimal capital structure: Theory and evidence. Journal of Finance, 39(3), 857–878.
• Brennan, M., & Kraft, H. (2018). Leaning against the wind: Debt financing in the face of adversity. Financial Management, 47(3), 485–518.
• Brennan, M., & Kraus, A. (1987). Efficient financing under asymmetric information. Journal of Finance, 42(5), 1225–1243.
• Butters, J. K. (1949). Federal income taxation and external vs. internal financing. Journal of Finance, 4(3), 197–205.
• Cadsby, C. B., Frank, M. Z., & Maksimovic, V. (1990). Pooling, separating, and semiseparating equilibria in financial markets: Some experimental evidence. Review of Financial Studies, 3(3), 315–342.
• Cadsby, C. B., Frank, M. Z., & Maksimovic, V. (1998). Equilibrium dominance in experimental financial markets. Review of Financial Studies, 11(1), 189–232.
• Chakraborty, A., & Yilmaz, B. (2011). Adverse selection and convertible bonds. Review of Economic Studies, 78(1), 148–175.
• Chang, X., & Dasgupta, S. (2009). Target behavior and financing: How conclusive is the evidence? Journal of Finance, 64(4), 1767–1796.
• Chirinko, R. S., & Singha, A. R. (2000). Testing static tradeoff against pecking order models of capital structure: A critical comment. Journal of Financial Economics, 58(3), 417–425.
• Constantinides, G. M., & Grundy, B. D. (1989). Optimal investment with stock repurchase and financing as signals. Review of Financial Studies, 2(4), 445–465.
• de Jong, A., Verbeek, M., & Verwijmeren, P. (2010). The impact of financing surpluses and large financing deficits on tests of the pecking order theory. Financial Management, 39(2), 733–756.
• DeAngelo, H., & Roll, R. (2015). How stable are corporate capital structures? Journal of Finance, 70(1), 373–418.
• Denis, D. J., & McKeon, S. B. (2018). Persistent operating losses and corporate financial policies. Working Paper, University of Pittsburgh and University of Oregon.
• Dybvig, P. H., & Zender, J. F. (1991). Capital structure and dividend irrelevance with asymmetric information. Review of Financial Studies, 4(1), 201–219.
• Eckbo, B. E., & Kisser, M. (2018). Tradeoff theory and leverage dynamics of high-frequency debt issuers. Working paper, Tuck School of Business, Hanover, NH.
• Eckbo, B. E., & Kisser, M. (2019). The leverage-profitability puzzle revisited. Working paper, Tuck School of Business, Hanover, NH.
• Eckbo, B. E., Masulis, R. W., & Norli, O. (2008). Security offerings. In B. Espen Eckbo (Ed.), Handbook of empirical corporate finance (pp. 233–373). Amsterdam, the Netherlands: Elsevier.
• Elsas, R., & Florysiak, D. (2015). Dynamic capital structure adjustment and the impact of fractional dependent variables. Journal of Financial and Quantitative Analysis, 50(5), 1105–1133.
• Fama, E. F., & French, K. R. (2002). Testing trade-off and pecking order predictions about dividends and debt. Review of Financial Studies, 15(1), 1–33.
• Fama, E. F., & French, K. R. (2005). Financing decisions: Who issues stock? Journal of Financial Economics, 76(3), 549–582.
• Fama, E. F., & French, K. R. (2012). Capital structure choices. Critical Finance Review, 1(1), 59–101.
• Farre-Mensa, J., Michaely, R., & Schmalz, M. (2014). Payout policy. Annual Review of Financial Economics, 6(1), 75–134.
• Faulkender, M., Flannery, M. J., Hankins, K. W., & Smith, J. M. (2012). Cash flows and leverage adjustments. Journal of Financial Economics, 103(3), 632–646.
• Fischer, E. O., Heinkel, R., & Zechner, J. (1989). Dynamic capital structure choice: Theory and tests. Journal of Finance, 44(1), 19–40.
• Flannery, M. J., & Rangan, K. P. (2006). Partial adjustment toward target capital structures. Journal of Financial Economics, 79(3), 469–506.
• Frank, M. Z., & Goyal, V. K. (2003). Testing the pecking order theory of capital structure. Journal of Financial Economics, 67(2), 217–248.
• Frank, M. Z., & Goyal, V. K. (2008). Trade-off and pecking order theories of debt. In B. Espen Eckbo (Ed.), Handbook of empirical corporate finance (pp. 135–202). Amsterdam, the Netherlands: Elsevier.
• Frank, M. Z., & Goyal, V. K. (2009). Capital structure decisions: Which factors are reliably important? Financial Management, 38(1), 1–37.
• Frank, M. Z., & Goyal, V. K. (2015). The profits–leverage puzzle revisited. Review of Finance, 19(6), 1415–1453.
• Frank, M. Z., & Sanati, A. (2019). Financing corporate growth. Working paper, University of Minnesota Twin Cities, Minneapolis.
• Frank, M. Z., & Shen, T. (2019). Corporate capital structure actions. Journal of Banking and Finance, 106, 384–402.
• Fulghieri, P., Garcia, D., & Hackbarth, D. (2020). Asymmetric information and the pecking (dis)order. Review of Finance.
• Gigerenzer, G., & Gaissmaier, W. (2011). Heuristic decision making. Annual Review of Psychology, 62(1), 451–482.
• Goldstein, R., Ju, N., & Leland, H. (2001). An EBIT-based model of dynamic capital structure. Journal of Business, 74(4), 483–512.
• Gomes, A., & Phillips, G. (2012). Why do public firms issue private and public securities? Journal of Financial Intermediation, 21(4), 619–658.
• Graham, J. R., Leary, M. T., & Roberts, M. R. (2015). A century of capital structure: The leveraging of corporate America. Journal of Financial Economics, 118(3), 658–683.
• Hackbarth, D. (2008). Managerial traits and capital structure decisions. Journal of Financial and Quantitative Analysis, 43(3), 843–881.
• Halov, N., & Heider, F. (2011). Capital structure, risk and asymmetric information. Quarterly Journal of Finance, 1(4), 767–809.
• Harris, M., & Raviv, A. (1991). The theory of capital structure. Journal of Finance, 46(1), 297–355.
• Heaton, J. B. (2002). Managerial optimism and corporate finance. Financial Management, 31(2), 33–45.
• Hennessy, C. A., & Whited, T. M. (2005). Debt dynamics. Journal of Finance, 60(3), 1129–1165.
• Hovakimian, A., & Li, G. (2010). Is the partial adjustment model a useful tool for capital structure research? Review of Finance, 16(3), 733–754.
• Hovakimian, A., Opler, T., & Titman, S. (2001). The debt-equity choice. Journal of Financial and Quantitative Analysis, 36(1), 1–24.
• Huang, R., & Ritter, J. R. (2009). Testing theories of capital structure and estimating the speed of adjustment. Journal of Financial and Quantitative Analysis, 44(2), 237–271.
• Huang, R., & Ritter, J. R. (2020). Corporate cash shortfalls and financing decisions. Review of Financial Studies, Forthcoming.
• Izhakian, Y., Yermack, D., & Zender, J. F. (2016). Ambiguity and the tradeoff theory of capital structure. Working paper.
• Jensen, M. C., & Meckling, W. H. (1976). Theory of the firm: Managerial behavior, agency costs and ownership structure. Journal of Financial Economics, 3(4), 305–360.
• Lambrecht, B. M., & Myers, S. C. (2012). A Lintner model of payout and managerial rents. Journal of Finance, 67(5), 1761–1810.
• Leary, M. T., & Roberts, M. R. (2005). Do firms rebalance their capital structures? Journal of Finance, 60(6), 2575–2619.
• Leary, M. T., & Roberts, M. R. (2010). The pecking order, debt capacity, and information asymmetry. Journal of Financial Economics, 95(3), 332–355.
• Lemmon, M. L., & Zender, J. F. (2010). Debt capacity and tests of capital structure theories. Journal of Financial and Quantitative Analysis, 45(5), 1161–1187.
• Lemmon, M. L., & Zender, J. F. (2019). Asymmetric information, debt capacity, and capital structure. Journal of Financial and Quantitative Analysis, 54(1), 31–59.
• Lintner, J. (1956). Distribution of incomes of corporations among dividends, retained earnings, and taxes. American Economic Review, 46(2), 97–113.
• Ljungqvist, L., & Sargent, T. J. (2018). Recursive macroeconomic theory (4th ed.). Cambridge, MA: MIT Press.
• Malenko, A., & Tsoy, A. (2019). Asymmetric information and security design under knightian uncertainty. Working paper, Boston College, Newton, MA.
• Malmendier, U., & Tate, G. (2005). CEO overconfidence and corporate investment. Journal of Finance, 60(6), 2661–2700.
• Malmendier, U., Tate, G., & Yan, J. (2011). Overconfidence and early-life experiences: The effect of managerial traits on corporate financial policies. Journal of Finance, 66(5), 1687–1733.
• Marcum, B., Martin, D. R., & Strickland, D. (2011). Are US firms stockpiling too much cash? Journal of Corporate Accounting & Finance, 23(1), 9–18.
• Modigliani, F., & Miller, M. (1958). The cost of capital, corporation finance and the theory of investment. American Economic Review, 48(3), 261–297.
• Modigliani, F., & Miller, M. H. (1963). Corporate income taxes and the cost of capital: A correction. American Economic Review, 53(3), 433–443.
• Morellec, E., & Schürhoff, N. (2011). Corporate investment and financing under asymmetric information. Journal of Financial Economics, 99(2), 262–288.
• Myers, S. C. (1984). The capital structure puzzle. Journal of Finance, 39(3), 574–592.
• Myers, S. C. (2003). Financing of corporations. In Handbook of the Economics of Finance (Vol. 1, pp. 215–253). Amsterdam, the Netherlands: Elsevier.
• Myers, S. C. (2015). Finance, theoretical and applied. Annual Review of Financial Economics, Vol. 7, 1–34.
• Myers, S. C., & Majluf, N. S. (1984). Corporate financing and investment decisions when firms have information that investors do not have. Journal of Financial Economics, 13(2), 187–221.
• Nachman, D. C., & Noe, T. H. (1994). Optimal design of securities under asymmetric information. Review of Financial Studies, 7(1), 1–44.
• Noe, T. H. (1988). Capital structure and signaling game equilibria. Review of Financial Studies, 1(4), 331–355.
• Phillips, G., & Sertsios, G. (2017). Financing and new product decisions of private and publicly traded firms. Review of Financial Studies, 30(5), 1744–1789.
• Rajan, R., & Zingales, L. (1995). What do we know about capital structure? Some evidence from international data. Journal of Finance, 50(5), 1421–1460.
• Shyam-Sunder, L., & Myers, S. C. (1999). Testing static tradeoff against pecking order models of capital structure. Journal of Financial Economics, 51(2), 219–244.
• Stiglitz, J. E. (1973). Taxation, corporate financial policy, and the cost of capital. Journal of Public Economics, 2(1), 1–34.
• Tirole, J. (2010). The theory of corporate finance. Princeton, NJ: Princeton University Press.

### Appendix: Variable Construction

 Symbol Variable Name Compustat Mnemonics $Inv$a Investment (capx + acq + ivch - siv - sppe - ivstch - ivaco)/$att−1$ [0.25ex] $Div$b Cash dividends dv/$att−1$ [0.25ex] $ΔWorkingCapital$c Change in working capital (recch + invch + apalch + txach + aoloch) $×$ (-1)/$att−1$ [0.25ex] $ΔCash$d Change in cash chech/$att−1$ [0.25ex] $Needs$ Needs $Inv+ΔWorkingCapital+Div+ΔCash$ [0.25ex] $CF$e Cash flow (ibc + dpc + xidoc + txdc + esubc + sppiv + fopo + exre)/$att−1$ [0.25ex] $Deficit$ Financing deficit $Needs$ - $CF$ [0.25ex] $ΔD$f Net debt issuance (dltis - dltr + dlcch + txbcof + fiao)/$att−1$ [0.25ex] $DI$ Debt issue (dltis + max[dlcch,0])/$att−1$ [0.25ex] $DR$ Debt repayment (dltr + min[dlcch,0] $×$ (-1))/$att−1$ [0.25ex] $pΔN$g Net equity issuance (sstk - prstkc)/$att−1$ [0.25ex] $EI$ Equity issue sstk/$att−1$ [0.25ex] $ER$ Equity repurchase prstkc/$att−1$ [0.25ex] $Financing$ Financing $CF+ΔD+pΔN$ [0.25ex] $ExtFin$ External financing $ΔD+pΔN$

Notes. The variables are constructed based on the cash flow statements provided in Compustat. Compustat mnemonics are included to identify the Compustat variable names. All of the variables are standardized by total assets (or “at”) at the end of the previous year.

a The numerator of $Inv$ is constructed as follows. For format code 7, it is capital expenditure (capx) + acquisitions (aqc) + increase in investment (ivch) – sale of investment (siv) – sale of property, plant, and equipment (sppe) – short-term investments-change (ivstch) – investing activities-other (ivaco). For format codes 1, 2, and 3, replace ivaco with the negative of use of funds-other (−1 $×$ fuseo).

b The numerator of $Div$ is cash dividends (dv).

c The numerator of $ΔWorkingCapital$ is constructed as follows. For format code 7, it is the negative of the sum of accounts receivable-decrease(increase) (recch), inventory-decrease(increase) (invch), accounts payable and accrued liabilities-increase(decrease) (apalch), income taxes-accrued-increase(decrease) (txach), and assets and liabilities-other (net change) (aoloch). For format code 1, the numerator of $ΔWorkingCapital$ is wcapch. For format codes 2 and 3, it is (−1) $×$ wcapc.

d The numerator of $ΔCash$ for format codes 2, 3, and 7 is cash and cash equivalents-increase(decrease) (chech). For format code 1, chech is recoded to 0.

e The numerator of $CF$ is constructed as follows. For format code 7, it is income before extraordinary items (ibc) + depreciation and amortization (dpc) + extraordinary items and discontinued operations (xidoc) + deferred taxes (txdc) + equity in net loss (earnings) (esubc) + sale of property, plant, and equipment and sale of investments-loss (gain) + funds from operations-other (fopo) + exchange rate effect (exre). For format codes 1, 2, or 3, replace exre with sources of funds-other (fsrco).

f The numerator of $ΔD$ is constructed as follows. For format code 7, it is long-term debt issuance (dltis) – long-term debt-reduction (dltr) + changes in current debt (dlcch) + excess tax benefits of stock options (txbcof) + financing activities-other (fiao). The items txbcof and fiao are both 0 for format codes 1, 2, and 3. Additionally, dlcch is recoded to zero for format code 1.

g The numerator of net equity issuance $(pΔN)$ is sale of common and preferred stock (sstk) – purchase of common and preferred stock (prstkc).