# Corporate Deleveraging and Financial Flexibility

Corporate Deleveraging and Financial Flexibility
DeAngelo, Harry;Gonçalves, Andrei S;Stulz, René M
2017-12-26 00:00:00
Abstract Most firms deleverage from their historical peak market-leverage (ML) ratios to near-zero ML, while also markedly increasing cash balances to high levels. Among 4,476 nonfinancial firms with five or more years of post-peak data, median ML is 0.543 at the peak and 0.026 at the later trough, with a six-year median time from peak to trough and with debt repayment and earnings retention accounting for 93.7% of the median peak-to-trough decline in ML. The findings support theories in which firms deleverage to restore ample financial flexibility and are difficult to reconcile with most firms having materially positive leverage targets. Received November 17, 2016; editorial decision November 9, 2017 by Editor David Denis. Authors have furnished an Internet Appendix, which is available on the Oxford University Press Web Site next to the link to the final published paper online. In capital structure theories that emphasize financial flexibility, firms choose financial policies to assure reliable and cost-efficient access to capital. Foundational elements of such theories are traceable to Donaldson (1961, pp. 105–6), Modigliani and Miller (1963, p. 442), and Myers and Majluf (1984), with recent developments discussed by Denis (2011). The key property of financial flexibility-based theories of capital structure is that firms use debt for transitory financing, with debt issuance ideally followed by eventual deleveraging to restore the option to borrow. Intuitively, firms tend to issue debt rather than equity to meet funding needs because of its lower marginal (flotation and asymmetric-information) costs. Consequently, firms choose to lever up at times, but prefer to avoid permanently high leverage because of the limited capacity to issue debt to meet new funding needs. Regardless of whether firms reach a leverage level with little remaining flexibility because they issued debt or because exogenous shocks reduced equity values, they have incentives to deleverage substantially so that they again have ample unused debt capacity they could tap in the future. Although chief financial officers (CFOs) say financial flexibility is the most important element of financial policy (Graham and Harvey 2001), the evidence in the literature raises doubts that highly levered firms systematically deleverage to restore ample flexibility. Specifically, five prior studies report modest average leverage decreases over long horizons for firms with high and/or recently increased leverage (Leary and Roberts 2005; Lemmon, Roberts, and Zender 2008; Harford, Klasa, and Walcott 2009; Denis and McKeon 2012; DeAngelo and Roll 2015). The largest deleveraging among these studies is found by Denis and McKeon, who report that the cross-firm average market-leverage ratio declines by 0.133 from almost 0.550 to just above 0.400 over the seven years after large increases in leverage. None of the studies finds a systematic tendency for the proactive deleveraging from high to conservative leverage that would be expected if financial flexibility were a critical driver of capital structure. In this paper, we find a strong tendency for firms to deleverage from high to conservative leverage, as predicted by financial flexibility-based theories of capital structure. Our conclusions differ from prior studies because we examine deleveraging on a firm-by-firm longitudinal basis rather than in terms of trends in cross-firm average leverage ratios. Our main analysis assesses deleveraging from the all-time high market-leverage (ML) ratio of each nonfinancial firm with post-peak data on Compustat to its subsequent ML trough, a period that takes six years for the median firm. We find that most of these firms deleverage from historical peak to near-zero ML, while simultaneously increasing cash balances to a level that is high in absolute terms and much higher than it was at peak ML. This deleveraging largely reflects managerial decisions to repay debt and retain earnings as opposed to exogenous shocks that drive stock-market prices up and ML ratios down. The fact that this large-scale deleveraging is typically accompanied by substantial increases in cash balances indicates that most sample firms are rebuilding financial flexibility generally, and not simply reducing leverage to low levels. Viewed most broadly, our findings are consistent with theories in which firms proactively deleverage to restore ample financial flexibility, and are difficult to reconcile with the idea that most firms have materially positive leverage targets. In most of this study, we analyze deleveraging from each firm’s all-time peak ML ratio to its later trough, but we also report the results of sensitivity checks that indicate that our main results continue to hold qualitatively for firms with recently increased and/or high (but not necessarily peak) ML. Existing theories that do not emphasize financial flexibility generally do not predict that firms will seek to deleverage from all-time peak (or high) ML to a conservative capital structure. For example, in traditional tax/distress cost trade-off theories, firms have tax incentives to maintain positive leverage ratios on a permanent basis. In such theories, firms would be expected to revert to a lower, but still materially positive, level of leverage after reaching peak ML. Our finding that firms tend to deleverage proactively from a typically quite high peak ML to near-zero ML favors flexibility-based theories, which indicate that firms have incentives to deleverage to conservative capital structures. Market leverage (ML $$=$$ debt/(debt $$+$$ equity market value)) is 0.543 at the all-time peak and 0.026 at the later trough for the median among 4,476 nonfinancial firms with at least five years of post-peak data on Compustat. About one-third (33.2%) of these firms pay off all debt they had at peak ML and well over half (60.3%) deleverage to negative net debt. The scale of these deleveraging episodes is also large in terms of book leverage (debt/total assets) and the net-debt ratio ((debt minus cash)/total assets), and when the sample includes firms with as few as one year of post-peak data on Compustat. For the latter all-inclusive sample of 9,866 firms, median ML is 0.491 at the peak and 0.088 at the later trough. Large-scale deleveraging is the norm across the spectrum of peak ML ratios, including among those firms that have the very highest peak levels of ML. Figure 1 shows that, in a sorting by peak-ML deciles of firms with five or more years of post-peak data, median ML at the post-peak trough is, for every decile group, well below the median ML ratio that prevailed at peak. Figure 1 View largeDownload slide Scale of deleveraging from peak market leverage (ML) to subsequent trough: Sample sorted by deciles of peak ML Market leverage (ML) is book debt divided by the sum of book debt and the market value of equity. Peak leverage is the maximum ML over a firm’s time in the sample. The subsequent trough is the lowest value of ML after the peak. The sample contains 4,476 nonfinancial firms that have five or more years of post-peak data on Compustat. Each of the ten decile groups accordingly contains 447 or 448 firms. For deciles 1 and 2, the median firm has zero debt at the post-peak trough, and so the figure shows a positive value for the median ML at the trough after peak only for deciles 3 to 10. Figure 1 View largeDownload slide Scale of deleveraging from peak market leverage (ML) to subsequent trough: Sample sorted by deciles of peak ML Market leverage (ML) is book debt divided by the sum of book debt and the market value of equity. Peak leverage is the maximum ML over a firm’s time in the sample. The subsequent trough is the lowest value of ML after the peak. The sample contains 4,476 nonfinancial firms that have five or more years of post-peak data on Compustat. Each of the ten decile groups accordingly contains 447 or 448 firms. For deciles 1 and 2, the median firm has zero debt at the post-peak trough, and so the figure shows a positive value for the median ML at the trough after peak only for deciles 3 to 10. We also find broad-based tendencies, evident in Figure 2, for firms to deleverage to zero-debt and negative-net-debt capital structures after having reached peak ML. The latter tendency reflects that deleveraging from peak ML is typically accompanied by decisions to increase cash holdings, with the median cash-to-total assets (Cash/TA) ratio increasing from 0.050 to 0.132 over the deleveraging episode for firms with at least five years of post-peak data. Among the 33.2% of these firms that repay all debt, median Cash/TA increases from 0.110 (when peak ML is 0.287) to 0.303 (when the ML trough of 0.000 is reached), thereby driving Net Debt/TA deeply negative. Figure 2 View largeDownload slide Percentage of firms that deleverage to zero debt and negative net debt capital structures: Sample sorted by deciles of peak market leverage Market leverage (ML) is book debt divided by the sum of book debt and the market value of equity. Peak leverage is the maximum ML over a firm’s time in the sample. The subsequent trough is the lowest value of ML after the peak. A negative-net-debt capital structure has a level of debt that is lower than the firm’s cash holdings. The sample contains 4,476 nonfinancial firms that have five or more years of post-peak data on Compustat. Each of the ten decile groups accordingly contains 447 or 448 firms. Figure 2 View largeDownload slide Percentage of firms that deleverage to zero debt and negative net debt capital structures: Sample sorted by deciles of peak market leverage Market leverage (ML) is book debt divided by the sum of book debt and the market value of equity. Peak leverage is the maximum ML over a firm’s time in the sample. The subsequent trough is the lowest value of ML after the peak. A negative-net-debt capital structure has a level of debt that is lower than the firm’s cash holdings. The sample contains 4,476 nonfinancial firms that have five or more years of post-peak data on Compustat. Each of the ten decile groups accordingly contains 447 or 448 firms. Although most firms deleverage from peak ML to a conservatively levered capital structure, a nontrivial minority do not, and financial distress is an important reason why. Almost 22% of firms are delisted due to distress in the year they reach peak ML or in the next four years. Firms delisted in the four years after peak, whether due to distress or acquisition, typically have ML ratios throughout their brief post-peak periods that are well above the ML ratios at the post-peak trough of firms that are not delisted. We gauge the component of deleveraging due to managerial decisions, including the direct effects of debt repayment and share issuance as well as the effect through the payout-policy channel of decisions to retain rather than pay out earnings. Debt repayment is the single most important endogenous element of deleveraging, accounting for 71.3% of the median peak-to-trough decline in ML for firms with five or more years of post-peak data. Together, debt repayment and earnings retention account for 93.7% of the peak-to-trough decline in ML for the median firm in this sample, whereas the inclusion of share issuances raises this percentage by only 2.8% to 96.5%. We focus on the time-series impact of cumulative earnings retention on ML ratios as opposed to the cross-sectional relation between leverage and current earnings, which many prior studies analyze (see Danis, Rettl, and Whited 2014 and the studies cited therein). Earnings retention increases market equity due to internally generated capital, which directly increases the denominator of the ML ratio and reduces ML for firms that have debt outstanding. The decision to pay out a given fraction of earnings is identical to the decision to retain one minus that fraction, and this endogenous payout/retention choice is not separable from the choice of leverage. Because payout/retention decisions are endogenous to managers, it is important to consider the full influence of what we call the payout-policy channel when seeking to gauge the extent to which managers shape leverage dynamics.1 We find that earnings retention makes a nontrivial contribution to the typical deleveraging episode, with especially strong contributions when ML is high and when firms increase their debt while reducing ML. On the other hand, although the well-known negative cross-sectional relation between leverage and profitability suggests that low leverage is often generated by deleveraging through earnings retention, low leverage in our sample is rarely the result of retention alone. Debt repayment is generally much more important than retention when firms deleverage to a conservative capital structure. While debt repayment is the most important direct contributor to deleveraging, its deleveraging impact is not fully independent of the new equity capital that firms obtain through earnings retention and share-issuance proceeds. The reason is that retention and issuance proceeds provide resources that can be used to repay debt and, for our sample firms, both forms of new equity are typically large relative to the amount of debt repaid (and to post-deleveraging cash balances). The implication is that internally generated and externally supplied new equity are both economically material indirect (funding-related) contributors to the deleveraging episodes we study. Firm leverage is both highly path dependent and closely linked to cash-balance policy. We find that a basic regression model with a firm’s peak ML and ML at the prior trough has roughly twice the power to explain ML at the post-peak trough than a model with industry ML, firm profitability, and other variables traditionally used to explain leverage, all evaluated at the post-peak trough ML. The R$$^{2}$$s are 36% and 19%, respectively. The difference becomes larger still as the R$$^{2}$$ increases to 53% when the simple model with ML at the peak and prior trough is augmented by information about whether a firm has had only a short time to deleverage, for example, due to distress-related delisting soon after peak, and about the level of cash the firm has accumulated at the post-peak trough. The implication is that the key to explaining whether ML at the outcome of deleveraging is relatively high or low is knowledge of (1) how high ML was at the peak and at the prior trough, (2) whether the firm has had time to work ML back down, and (3) how much the firm has rebuilt the cash-balance component of financial flexibility. These findings on cross-firm variation in deleveraging are robust to inclusion of a variety of other explanatory variables. The most important finding in these robustness checks is that proactive increases in ML (defined as in Denis and McKeon 2012) in the year that peak leverage is reached imply a statistically significant, but economically immaterial, difference in subsequent deleveraging outcomes. Our finding that the typical firm deleverages from all-time peak to a near-zero ML ratio differs sharply from the relatively muted leverage reductions reported in prior studies that examine deleveraging over long horizons. An important reason for the large difference in our findings about the size of deleveraging is that prior studies do not use a longitudinal approach. Instead, they first calculate average leverage ratios for a set of firms at each point in (event) time, and then assess the extent of reductions (over event time) in the cross-firm average leverage ratio. DeAngelo and Roll (2015, p. 392) point out that analyzing trends in cross-sectional average leverage ratios can be misleading because large-sample averaging masks the substantial time-series volatility in the leverage of most firms that they document. For the study of deleveraging, the problem with comparisons of event-time averages is bias, not the masking of volatility. We show that such comparisons underestimate the size of the typical firm’s deleveraging when, as is true in our data, the length of deleveraging episodes differs across firms and the leverage ratios of many firms do not stabilize near their post-peak leverage troughs. We use the longitudinal approach to conduct robustness checks of our main findings when firms deleverage after their ML ratios increase markedly, but not necessarily to their all-time peak levels. In our robustness checks, we also examine deleveraging from leverage peaks measured in book-value terms and after book leverage increases markedly, but not necessarily to all-time peak levels. Our main findings are qualitatively unchanged in this robustness analysis. We discuss the implications of our findings for flexibility-based and alternative theories of financial policy in the last section of the paper. 1. Sample Construction and a First Look at Deleveraging This section describes our sampling procedure and presents evidence on year-to-year changes in leverage that motivates our longitudinal long-run perspective for studying deleveraging. 1.1 Sample construction We begin by identifying 15,703 publicly held nonfinancial firms that are in the CRSP/Compustat file at some point over 1950 to 2012. Firms in this sample are required to be incorporated in the US and to have CRSP security codes of 10 or 11 and SIC codes outside the ranges 4900 to 4949 (utilities) and 6000 to 6999 (financials). Firm-year observations are included if they have nonmissing values on Compustat of the market value of equity (common stock) and the book values of total assets and cash balances. Total debt is the sum of the book values of short- and long-term debt, and a firm-year observation is included only if at least one of these two debt components is nonmissing, with the other component set to zero if it is missing. We arrive at our baseline sample of 14,196 firms after exclusion of 962 firms with only one year of data and 545 firms that always have zero debt while on Compustat. Of the 14,196 firms in the baseline sample, 9,866 firms have data on Compustat for at least one year after reaching their historical peak market-leverage ratio. These firms are the central focus of our deleveraging analysis. For the other 4,330 firms, there are no post-peak data that would allow us to gauge the nature and extent of deleveraging. The latter firms enter our analysis in Section 4, which investigates the link between high leverage and early sample exits due to financial distress and mergers. In constructing an appropriate sample for our study, a potential problem arises because many firms have just a few years of data on Compustat, yet there is good reason to think that deleveraging often takes seven years or more (Denis and McKeon 2012). The concern is that many firms do not remain on Compustat long enough for us to be able to observe their full deleveraging, so that the deleveraging that we detect is attenuated for these firms. DeAngelo and Roll (2015) show that Compustat’s “short-sample” property can mask large leverage instability because of the inclusion of many firms with attenuated measures of leverage changes. For our study, the important concern is that any sampling rule that is tilted toward firms with a limited number of years of data can inject a downward bias into estimates of the magnitude of corporate deleveraging. This problem is potentially important in all Compustat-based samples, including our baseline sample where only 4,476 (45.4%) of the 9,866 firms with observable deleveraging episodes have five or more years of post-peak data on Compustat. This potential “short-sample” bias suggests that a sample-inclusion requirement that firms have data available for an extended period may be essential for an informative analysis of deleveraging. On the other hand, such a sampling requirement has its own possible bias, namely that firms that have survived for an extended period may differ in empirically relevant ways from those with limited data available. For example, a plausible worry is that, if we require firms to have data available for an extended period, we would exclude many firms that reach a high ML ratio and are then delisted early due to financial distress. Firms that reach a high ML ratio because distress has eroded equity value likely have the extent of their deleveraging attenuated as managerial attempts to reduce ML are thwarted by the distress itself. The general concern, therefore, is that a sample restricted to firms with many years of data would be informative only about deleveraging by successful firms. It would fail to present a complete picture by materially under-representing firms whose financial troubles led them to disappear from the public arena before they had logged enough years of data to qualify for such a sample. We address these issues by analyzing subsets of the baseline sample in which firms have successively larger numbers of years of data available. We also gauge the extent of attenuated deleveraging associated with early sample exits, for example, due to delisting by distressed firms. This approach enables us to make empirically informative statements about deleveraging episodes conditional on the amount of time firms have leverage data in the public domain. This approach is appropriate when studying leverage dynamics because firm survival is necessary for researchers to have the data to gauge leverage changes over time. 1.2 Annual deleveraging propensities We focus throughout the paper on deleveraging in terms of market-leverage ratios, but we also report book leverage as well as cash and net-debt ratios when relevant. Market leverage (ML) is the book value of total debt divided by book debt plus the market value of equity. Book leverage (BL or Debt/TA) is total debt divided by total assets in book terms. The cash ratio (Cash/TA) is cash plus marketable securities divided by total assets. The net-debt ratio (Net Debt/TA) is Debt/TA minus Cash/TA. Table 1 reports annual leverage changes using our baseline sample and adding back firms with zero debt in all years. For this study, the most important regularity in the table is that, when leverage is high, it tends to decrease in the next year, with the typical reduction modest in size. Specifically, when ML, BL, and Net Debt/TA exceed 0.500, there is roughly a 60.0% probability of a leverage decrease in the next year, with each leverage measure showing a median change around $$-0.020$$ (row 1). When these leverage measures exceed 0.400 or 0.300, the probability of a leverage decrease is lower and the median changes remain negative, but are closer to zero (rows 2 and 3). The tendency for leverage to decrease is weaker at lower levels of ML, BL, and Net Debt/TA and there is a slight tendency for ML and Net Debt/TA to increase when they are currently low (rows 4 to 6). The overall pattern of year-over-year leverage changes is consistent with the weak mean reversion reported in prior studies.2 Table 1 Deleveraging propensities: Annual changes in leverage as a function of the beginning-of-year level of leverage Market leverage Book leverage Net Debt/TA Beginning-of-year ratio (ML or BL or Net Debt/TA) Probability of annual decrease Median annual change Probability of annual decrease Median annual change Probability of annual decrease Median annual change 1. 0.500 $$<$$ Leverage 57.8% –0.022 60.4% –0.019 61.0% –0.022 2. 0.400 $$<$$ Leverage $$\leqslant$$ 0.500 53.6% –0.012 58.6% –0.014 59.1% –0.016 3. 0.300 $$<$$ Leverage $$\leqslant$$ 0.400 51.4% –0.004 56.6% –0.010 56.2% –0.011 4. 0.200 $$<$$ Leverage $$\leqslant$$ 0.300 48.3% 0.004 53.9% –0.005 52.7% –0.005 5. 0.100 $$<$$ Leverage $$\leqslant$$ 0.200 45.9% 0.008 51.7% –0.002 49.2% 0.001 6. 0.000 $$<$$ Leverage $$\leqslant$$ 0.100 45.9% 0.002 53.5% –0.001 45.3% 0.009 7. All leverage $$>$$ 0.000 49.9% 0.000 55.0% –0.003 – – 8. All including BL $$=$$ ML $$=$$ 0.000 43.6% 0.000 47.8% 0.000 – – 9. All Net Debt/TA $$\geqslant$$ 0.000 – – – – 52.5% –0.005 10. All Net Debt/TA $$<$$ 0.000 – – – – 38.3% 0.029 Market leverage Book leverage Net Debt/TA Beginning-of-year ratio (ML or BL or Net Debt/TA) Probability of annual decrease Median annual change Probability of annual decrease Median annual change Probability of annual decrease Median annual change 1. 0.500 $$<$$ Leverage 57.8% –0.022 60.4% –0.019 61.0% –0.022 2. 0.400 $$<$$ Leverage $$\leqslant$$ 0.500 53.6% –0.012 58.6% –0.014 59.1% –0.016 3. 0.300 $$<$$ Leverage $$\leqslant$$ 0.400 51.4% –0.004 56.6% –0.010 56.2% –0.011 4. 0.200 $$<$$ Leverage $$\leqslant$$ 0.300 48.3% 0.004 53.9% –0.005 52.7% –0.005 5. 0.100 $$<$$ Leverage $$\leqslant$$ 0.200 45.9% 0.008 51.7% –0.002 49.2% 0.001 6. 0.000 $$<$$ Leverage $$\leqslant$$ 0.100 45.9% 0.002 53.5% –0.001 45.3% 0.009 7. All leverage $$>$$ 0.000 49.9% 0.000 55.0% –0.003 – – 8. All including BL $$=$$ ML $$=$$ 0.000 43.6% 0.000 47.8% 0.000 – – 9. All Net Debt/TA $$\geqslant$$ 0.000 – – – – 52.5% –0.005 10. All Net Debt/TA $$<$$ 0.000 – – – – 38.3% 0.029 Market leverage (ML) is the ratio of the book value of debt to the sum of the book value of debt and the market value of equity. Book leverage (BL or Debt/TA) is the ratio of the book value of total (short-term plus long-term) debt to the book value of total assets. The net-debt ratio (NDR or Net Debt/TA) equals debt minus cash, divided by total assets. The full sample contains 14,741 nonfinancial firms with a total of 171,010 firm-year observations in the CRSP/Compustat file over 1950–2012. The number of firms here exceeds the number in the baseline sample because here we do not exclude firms that have the same ML ratio in all years (as we do in the baseline sample). The data on changes in book leverage and net-debt exclude firm-year observations with Debt/TA above 1.000. Table 1 Deleveraging propensities: Annual changes in leverage as a function of the beginning-of-year level of leverage Market leverage Book leverage Net Debt/TA Beginning-of-year ratio (ML or BL or Net Debt/TA) Probability of annual decrease Median annual change Probability of annual decrease Median annual change Probability of annual decrease Median annual change 1. 0.500 $$<$$ Leverage 57.8% –0.022 60.4% –0.019 61.0% –0.022 2. 0.400 $$<$$ Leverage $$\leqslant$$ 0.500 53.6% –0.012 58.6% –0.014 59.1% –0.016 3. 0.300 $$<$$ Leverage $$\leqslant$$ 0.400 51.4% –0.004 56.6% –0.010 56.2% –0.011 4. 0.200 $$<$$ Leverage $$\leqslant$$ 0.300 48.3% 0.004 53.9% –0.005 52.7% –0.005 5. 0.100 $$<$$ Leverage $$\leqslant$$ 0.200 45.9% 0.008 51.7% –0.002 49.2% 0.001 6. 0.000 $$<$$ Leverage $$\leqslant$$ 0.100 45.9% 0.002 53.5% –0.001 45.3% 0.009 7. All leverage $$>$$ 0.000 49.9% 0.000 55.0% –0.003 – – 8. All including BL $$=$$ ML $$=$$ 0.000 43.6% 0.000 47.8% 0.000 – – 9. All Net Debt/TA $$\geqslant$$ 0.000 – – – – 52.5% –0.005 10. All Net Debt/TA $$<$$ 0.000 – – – – 38.3% 0.029 Market leverage Book leverage Net Debt/TA Beginning-of-year ratio (ML or BL or Net Debt/TA) Probability of annual decrease Median annual change Probability of annual decrease Median annual change Probability of annual decrease Median annual change 1. 0.500 $$<$$ Leverage 57.8% –0.022 60.4% –0.019 61.0% –0.022 2. 0.400 $$<$$ Leverage $$\leqslant$$ 0.500 53.6% –0.012 58.6% –0.014 59.1% –0.016 3. 0.300 $$<$$ Leverage $$\leqslant$$ 0.400 51.4% –0.004 56.6% –0.010 56.2% –0.011 4. 0.200 $$<$$ Leverage $$\leqslant$$ 0.300 48.3% 0.004 53.9% –0.005 52.7% –0.005 5. 0.100 $$<$$ Leverage $$\leqslant$$ 0.200 45.9% 0.008 51.7% –0.002 49.2% 0.001 6. 0.000 $$<$$ Leverage $$\leqslant$$ 0.100 45.9% 0.002 53.5% –0.001 45.3% 0.009 7. All leverage $$>$$ 0.000 49.9% 0.000 55.0% –0.003 – – 8. All including BL $$=$$ ML $$=$$ 0.000 43.6% 0.000 47.8% 0.000 – – 9. All Net Debt/TA $$\geqslant$$ 0.000 – – – – 52.5% –0.005 10. All Net Debt/TA $$<$$ 0.000 – – – – 38.3% 0.029 Market leverage (ML) is the ratio of the book value of debt to the sum of the book value of debt and the market value of equity. Book leverage (BL or Debt/TA) is the ratio of the book value of total (short-term plus long-term) debt to the book value of total assets. The net-debt ratio (NDR or Net Debt/TA) equals debt minus cash, divided by total assets. The full sample contains 14,741 nonfinancial firms with a total of 171,010 firm-year observations in the CRSP/Compustat file over 1950–2012. The number of firms here exceeds the number in the baseline sample because here we do not exclude firms that have the same ML ratio in all years (as we do in the baseline sample). The data on changes in book leverage and net-debt exclude firm-year observations with Debt/TA above 1.000. 1.3 Long-horizon longitudinal analysis of deleveraging These findings suggest that large-scale deleveraging tends to play out in small-to-moderate steps over multiyear horizons. This in turn suggests—and Section 2 strongly confirms—that focusing on year-to-year leverage changes tends to miss their cumulative effect at firms that are going through material deleveraging, and therefore often fails to identify such episodes. We accordingly adopt a long-horizon longitudinal approach that, for each sample firm, analyzes deleveraging from all-time peak ML to subsequent trough. We have a total of 14,196 observations, one for each firm in the baseline sample. We focus primarily on the 9,866 firms that have post-peak data available on Compustat. In Section 4’s analysis of attenuated deleveraging, we also consider the 4,330 firms that have no post-peak data, for example, due to distress-related delisting in the peak year. This broad-based sample includes (a) firms that attain peak ML proactively, (b) firms that attain peak ML as a result of exogenous shocks that decrease the market value of equity, and (c) all cases, regardless of the size of the ML increase in the year a firm reaches peak. Conditions (a) and (b) imply that our sample contains many observations that are not included in the sample of Denis and McKeon (2012), who study deleveraging by firms that increase leverage by large amounts to a high (but not necessarily peak) level. At the same time, our sample includes 3,000 firms whose movements to peak ML satisfy their definition of a proactive ML increase. Sections 5 and 6 analyze the difference in deleveraging outcomes when peak is attained because managers chose to lever up rather than because exogenous shocks increase ML. While the longitudinal approach reveals substantial heterogeneity across firms in the time between all-time peak ML and subsequent trough, we also find that a large majority of these deleveraging episodes play out over a relatively compact period of a decade or less. The time from peak to trough is 10 years or less for 90.2% of the 9,866 firms with at least one year of post-peak data available on Compustat and for 78.4% of the 4,476 firms with at least five years of post-peak data available. 2. Deleveraging and the Restoration of Financial Flexibility Table 2 documents that deleveraging from all-time peak market leverage (ML) to subsequent ML trough transforms the typical firm from a capital structure with far more debt than cash to one with ample financial flexibility in terms of both low leverage and much higher cash balances. The first column of the table reports results for our all-inclusive baseline sample, that is, for all 9,866 firms with at least one year of post-peak data on Compustat. Moving sequentially from left to right, the remaining columns report results for subsets of the baseline sample with firms that have a minimum of two years of post-peak data (second column) up to a minimum of ten years of post-peak data on Compustat (far-right column). Table 2 Deleveraging episodes: Market leverage (ML) and related financial ratios at the ML peak and subsequent ML trough, with the baseline sample partitioned by the minimum number of years of post-peak data available on Compustat Minimum number of years of data available after the market leverage (ML) peak: $$\geqslant 1$$ $$\geqslant 2$$ $$\geqslant 3$$ $$\geqslant 4$$ $$\geqslant 5$$ $$\geqslant 6$$ $$\geqslant 7$$ $$\geqslant 8$$ $$\geqslant 9$$ $$\geqslant 10$$ 1. Median ML at peak 0.491 0.509 0.526 0.535 0.543 0.552 0.557 0.562 0.571 0.570 2. Median ML at trough after peak 0.088 0.062 0.048 0.040 0.026 0.023 0.020 0.018 0.017 0.016 3. Percentage of firms that have $$\qquad$$ zero debt at ML trough after peak 22.8% 26.7% 28.9% 30.5% 33.2% 34.0% 34.7% 35.3% 35.8% 36.0% $$\qquad$$ negative net debt at ML trough after peak 49.1% 53.6% 55.9% 57.6% 60.3% 61.4% 62.4% 63.4% 63.8% 64.6% 4. Median Cash/TA at ML peak 0.056 0.055 0.054 0.052 0.050 0.049 0.048 0.047 0.047 0.046 5. Median Cash/TA at post-peak ML trough 0.109 0.120 0.126 0.130 0.132 0.134 0.136 0.136 0.136 0.137 6. Median Net Debt/TA at ML peak 0.283 0.286 0.288 0.291 0.296 0.299 0.302 0.304 0.310 0.310 7. Median Net Debt/TA at post-peak ML trough 0.007 –0.023 –0.040 –0.052 –0.067 –0.072 –0.076 –0.079 –0.081 –0.084 8. Median book leverage (BL) at ML peak 0.354 0.354 0.356 0.358 0.359 0.360 0.361 0.362 0.365 0.363 9. Median BL at post-peak ML trough 0.121 0.090 0.076 0.063 0.044 0.036 0.032 0.028 0.026 0.024 10. Median peak-to-trough decline in ML –0.244 –0.305 –0.339 –0.365 –0.395 –0.414 –0.427 –0.438 –0.451 –0.462 11. Median years from ML peak to trough 2 3 4 5 6 6 7 8 9 9 12. Number of firms 9,866 7,801 6,529 5,547 4,476 3,954 3,467 3,075 2,756 2,462 Minimum number of years of data available after the market leverage (ML) peak: $$\geqslant 1$$ $$\geqslant 2$$ $$\geqslant 3$$ $$\geqslant 4$$ $$\geqslant 5$$ $$\geqslant 6$$ $$\geqslant 7$$ $$\geqslant 8$$ $$\geqslant 9$$ $$\geqslant 10$$ 1. Median ML at peak 0.491 0.509 0.526 0.535 0.543 0.552 0.557 0.562 0.571 0.570 2. Median ML at trough after peak 0.088 0.062 0.048 0.040 0.026 0.023 0.020 0.018 0.017 0.016 3. Percentage of firms that have $$\qquad$$ zero debt at ML trough after peak 22.8% 26.7% 28.9% 30.5% 33.2% 34.0% 34.7% 35.3% 35.8% 36.0% $$\qquad$$ negative net debt at ML trough after peak 49.1% 53.6% 55.9% 57.6% 60.3% 61.4% 62.4% 63.4% 63.8% 64.6% 4. Median Cash/TA at ML peak 0.056 0.055 0.054 0.052 0.050 0.049 0.048 0.047 0.047 0.046 5. Median Cash/TA at post-peak ML trough 0.109 0.120 0.126 0.130 0.132 0.134 0.136 0.136 0.136 0.137 6. Median Net Debt/TA at ML peak 0.283 0.286 0.288 0.291 0.296 0.299 0.302 0.304 0.310 0.310 7. Median Net Debt/TA at post-peak ML trough 0.007 –0.023 –0.040 –0.052 –0.067 –0.072 –0.076 –0.079 –0.081 –0.084 8. Median book leverage (BL) at ML peak 0.354 0.354 0.356 0.358 0.359 0.360 0.361 0.362 0.365 0.363 9. Median BL at post-peak ML trough 0.121 0.090 0.076 0.063 0.044 0.036 0.032 0.028 0.026 0.024 10. Median peak-to-trough decline in ML –0.244 –0.305 –0.339 –0.365 –0.395 –0.414 –0.427 –0.438 –0.451 –0.462 11. Median years from ML peak to trough 2 3 4 5 6 6 7 8 9 9 12. Number of firms 9,866 7,801 6,529 5,547 4,476 3,954 3,467 3,075 2,756 2,462 Market leverage (ML) is the ratio of the book value of debt to the sum of the book value of debt and the market value of equity. Peak leverage is the maximum value of ML over a firm’s entire time in the sample. The trough after peak is the lowest value of a firm’s ML that comes subsequent to its peak. When a firm has multiple post-peak years with the same minimum value of ML, we take the earliest such year to be the date of the post-peak trough. Book leverage (BL) is the ratio of the book value of debt to the book value of total assets. Net Debt/TA is the book value of debt minus the sum of cash and marketable securities, divided by total assets. Results for the full baseline sample are in the first column. Successive columns report results for subsets of the baseline sample that exclude firms with a small number of years of post-peak data available on Compustat. For example, the column labeled “$$\geqslant 2$$” excludes the 2,065 firms (9,866 minus 7,801, per row 12) that have exactly one year of post-peak data available. Table 2 Deleveraging episodes: Market leverage (ML) and related financial ratios at the ML peak and subsequent ML trough, with the baseline sample partitioned by the minimum number of years of post-peak data available on Compustat Minimum number of years of data available after the market leverage (ML) peak: $$\geqslant 1$$ $$\geqslant 2$$ $$\geqslant 3$$ $$\geqslant 4$$ $$\geqslant 5$$ $$\geqslant 6$$ $$\geqslant 7$$ $$\geqslant 8$$ $$\geqslant 9$$ $$\geqslant 10$$ 1. Median ML at peak 0.491 0.509 0.526 0.535 0.543 0.552 0.557 0.562 0.571 0.570 2. Median ML at trough after peak 0.088 0.062 0.048 0.040 0.026 0.023 0.020 0.018 0.017 0.016 3. Percentage of firms that have $$\qquad$$ zero debt at ML trough after peak 22.8% 26.7% 28.9% 30.5% 33.2% 34.0% 34.7% 35.3% 35.8% 36.0% $$\qquad$$ negative net debt at ML trough after peak 49.1% 53.6% 55.9% 57.6% 60.3% 61.4% 62.4% 63.4% 63.8% 64.6% 4. Median Cash/TA at ML peak 0.056 0.055 0.054 0.052 0.050 0.049 0.048 0.047 0.047 0.046 5. Median Cash/TA at post-peak ML trough 0.109 0.120 0.126 0.130 0.132 0.134 0.136 0.136 0.136 0.137 6. Median Net Debt/TA at ML peak 0.283 0.286 0.288 0.291 0.296 0.299 0.302 0.304 0.310 0.310 7. Median Net Debt/TA at post-peak ML trough 0.007 –0.023 –0.040 –0.052 –0.067 –0.072 –0.076 –0.079 –0.081 –0.084 8. Median book leverage (BL) at ML peak 0.354 0.354 0.356 0.358 0.359 0.360 0.361 0.362 0.365 0.363 9. Median BL at post-peak ML trough 0.121 0.090 0.076 0.063 0.044 0.036 0.032 0.028 0.026 0.024 10. Median peak-to-trough decline in ML –0.244 –0.305 –0.339 –0.365 –0.395 –0.414 –0.427 –0.438 –0.451 –0.462 11. Median years from ML peak to trough 2 3 4 5 6 6 7 8 9 9 12. Number of firms 9,866 7,801 6,529 5,547 4,476 3,954 3,467 3,075 2,756 2,462 Minimum number of years of data available after the market leverage (ML) peak: $$\geqslant 1$$ $$\geqslant 2$$ $$\geqslant 3$$ $$\geqslant 4$$ $$\geqslant 5$$ $$\geqslant 6$$ $$\geqslant 7$$ $$\geqslant 8$$ $$\geqslant 9$$ $$\geqslant 10$$ 1. Median ML at peak 0.491 0.509 0.526 0.535 0.543 0.552 0.557 0.562 0.571 0.570 2. Median ML at trough after peak 0.088 0.062 0.048 0.040 0.026 0.023 0.020 0.018 0.017 0.016 3. Percentage of firms that have $$\qquad$$ zero debt at ML trough after peak 22.8% 26.7% 28.9% 30.5% 33.2% 34.0% 34.7% 35.3% 35.8% 36.0% $$\qquad$$ negative net debt at ML trough after peak 49.1% 53.6% 55.9% 57.6% 60.3% 61.4% 62.4% 63.4% 63.8% 64.6% 4. Median Cash/TA at ML peak 0.056 0.055 0.054 0.052 0.050 0.049 0.048 0.047 0.047 0.046 5. Median Cash/TA at post-peak ML trough 0.109 0.120 0.126 0.130 0.132 0.134 0.136 0.136 0.136 0.137 6. Median Net Debt/TA at ML peak 0.283 0.286 0.288 0.291 0.296 0.299 0.302 0.304 0.310 0.310 7. Median Net Debt/TA at post-peak ML trough 0.007 –0.023 –0.040 –0.052 –0.067 –0.072 –0.076 –0.079 –0.081 –0.084 8. Median book leverage (BL) at ML peak 0.354 0.354 0.356 0.358 0.359 0.360 0.361 0.362 0.365 0.363 9. Median BL at post-peak ML trough 0.121 0.090 0.076 0.063 0.044 0.036 0.032 0.028 0.026 0.024 10. Median peak-to-trough decline in ML –0.244 –0.305 –0.339 –0.365 –0.395 –0.414 –0.427 –0.438 –0.451 –0.462 11. Median years from ML peak to trough 2 3 4 5 6 6 7 8 9 9 12. Number of firms 9,866 7,801 6,529 5,547 4,476 3,954 3,467 3,075 2,756 2,462 Market leverage (ML) is the ratio of the book value of debt to the sum of the book value of debt and the market value of equity. Peak leverage is the maximum value of ML over a firm’s entire time in the sample. The trough after peak is the lowest value of a firm’s ML that comes subsequent to its peak. When a firm has multiple post-peak years with the same minimum value of ML, we take the earliest such year to be the date of the post-peak trough. Book leverage (BL) is the ratio of the book value of debt to the book value of total assets. Net Debt/TA is the book value of debt minus the sum of cash and marketable securities, divided by total assets. Results for the full baseline sample are in the first column. Successive columns report results for subsets of the baseline sample that exclude firms with a small number of years of post-peak data available on Compustat. For example, the column labeled “$$\geqslant 2$$” excludes the 2,065 firms (9,866 minus 7,801, per row 12) that have exactly one year of post-peak data available. Scanning across the columns in Table 2 reveals that large decreases in ML and substantial increases in Cash/TA ratios characterize the baseline sample and all subsets thereof, with the typical scale of deleveraging greater when the sample excludes firms with just a few years of post-peak data. Median ML is 0.543 at the peak and 0.026 at the later trough for the sample with at least five years of post-peak data, whereas the comparable figures are 0.491 and 0.088 for the baseline sample, which includes many firms with few years of post-peak data available (rows 1 and 2). Remarkably, 33.2% of the former firms and 22.8% of the firms in the baseline sample pay off all debt, whereas 60.3% and 49.1% of firms in the two samples deleverage to a negative net debt capital structure (row 3). The pervasive deleveraging to a negative net debt capital structure reflects the fact that firms typically increase cash balances by a nontrivial amount while deleveraging from peak ML. Among firms with five or more years of post-peak data, the median Cash/TA ratio almost triples from 0.050 at the ML peak to 0.132 at the later trough (rows 4 and 5). In the baseline sample, the analogous figures are 0.056 and 0.109 for a near doubling of Cash/TA. For all samples in the columns of Table 2, a comparison of peak and trough median ML (rows 1 and 2) with median Cash/TA (rows 4 and 5) is strongly suggestive of a counter-cyclical relation between leverage and cash balances at the individual-firm level. As with market leverage, the book leverage (BL) and Net Debt/TA ratios in Table 2 also indicate that our sample firms typically deleverage to conservative capital structures (rows 7 and 9). The median peak-to-trough change in ML is $$-0.244$$ for the baseline sample and $$-0.395$$ for firms with five or more years of post-peak data (row 10). The difference reflects the fact that the baseline sample contains quite a few firms that are delisted shortly after peak either because of financial distress or because they are acquired soon after attaining peak (see Section 4). Since these firms have little or no time to deleverage, their inclusion in the baseline sample naturally dampens the median decline in ML relative to the median among firms that have a minimum of five years to deleverage. Among firms with five or more years of post-peak data, the median deleveraging takes six years (row 11, Table 2), which is near the seven-year deleveraging horizon analyzed by Denis and McKeon (2012). In contrast, for our baseline sample, the median time from ML peak to trough is only two years. This brief deleveraging time is quite misleading because it is mechanically driven by the fact that the baseline sample has many firms with just a few years of post-peak data. More than half (54.6%) of firms in the baseline sample have four or fewer years of post-peak data, while about one-third (33.8%) have one or two years of data (row 12). It is therefore impossible for the median deleveraging time to be longer than four years, and difficult for it to be longer than two years. A closely related important regularity, evident in Figure 3, is that the scale of deleveraging is more muted among firms with just a few years of post-peak data. Panel A of the figure plots ML medians at the peak and the later trough (previously reported in Table 2) juxtaposed for each sample against the ML median at the trough that prevailed before the peak. Panel B of the figure reports ML medians at the pre-peak trough, the peak, and the post-peak trough for the incremental sets of firms that are excluded from the baseline sample as we move step by step to the right in Table 2 and in panel A of Figure 3, that is, for firms with exactly one year of post-peak data, exactly two years of post-peak data, and so on. Figure 3 View largeDownload slide Median market leverage (ML) at the peak and at the troughs before and after the peak: Sample sorted by the number of years of post-peak data on Compustat Market leverage (ML) is the ratio of the book value of debt to the sum of the book value of debt and the market value of equity. Peak leverage is the maximum value of ML over a firm’s entire time in the sample. The ML trough after (before) peak is the lowest ML value that comes after (before) the ML peak. Results for the full baseline sample are reported at the far left in panel A. The baseline sample has 9,866 firms with at least one year of post-peak data on Compustat. Sample sizes for the other subsample entries in panel A are in row 13 of Table 2. Sample sizes for panel B can be calculated by taking differences in the row 13 entries for adjacent columns in Table 2. Figure 3 View largeDownload slide Median market leverage (ML) at the peak and at the troughs before and after the peak: Sample sorted by the number of years of post-peak data on Compustat Market leverage (ML) is the ratio of the book value of debt to the sum of the book value of debt and the market value of equity. Peak leverage is the maximum value of ML over a firm’s entire time in the sample. The ML trough after (before) peak is the lowest ML value that comes after (before) the ML peak. Results for the full baseline sample are reported at the far left in panel A. The baseline sample has 9,866 firms with at least one year of post-peak data on Compustat. Sample sizes for the other subsample entries in panel A are in row 13 of Table 2. Sample sizes for panel B can be calculated by taking differences in the row 13 entries for adjacent columns in Table 2. Panel B of Figure 3 indicates that firms with exactly one year of post-peak data reduce leverage to a median ML above 0.200, while firms with exactly two, three, or four years of data deleverage to median ML ratios above 0.100. For each of these short-horizon samples, median ML at the post-peak trough is far higher than the median ML that had prevailed at the trough before the peak. In sharp contrast, among firms with five or more years of data, the median ML ratios at the post-peak trough are well below 0.100 and close to the median ML that prevailed at the pre-peak trough. Why do firms with just a few years of data tend to deleverage by smaller amounts than firms with more years of post-peak data on Compustat? One reason is that deleveraging typically plays out over multiple years, not through a one-time rebalancing of capital structure. Consequently, for many firms that have just a few years of post-peak data on Compustat, we can observe only a truncated portion of the full deleveraging. Another reason is that quite a few firms are delisted due to financial distress soon after reaching peak ML and, as prior research has documented, distressed firms commonly have difficulties restructuring their finances to obtain greater breathing room. We return to this issue in Section 4. Finally, we note that for the baseline sample and all subsamples in panel A of Figure 3, median ML at the trough before the peak is close to zero, just as it is at the trough after the peak. Symmetrically, median Cash/TA ratios are considerably higher at both troughs than at peak ML (panel A of Figure 4). More than half of the firms in the baseline sample have negative net debt at the trough before the peak, as do roughly half of the firms in all subsamples (panel B of Figure 4). In short, substantial financial flexibility—low ML coupled with high Cash/TA—is the norm at the ML troughs both before and after peak ML. Figure 4 View largeDownload slide Median cash-balance ratios and incidence of negative net debt capital structures at the troughs before and after peak market leverage Market leverage (ML) is the ratio of the book value of debt to the sum of the book value of debt and the market value of equity. Peak leverage is the maximum value of ML over a firm’s entire time in the sample. The ML trough after (before) peak is the lowest ML value that comes after (before) the ML peak. Cash/TA is the ratio of cash plus marketable securities to total assets. A negative net debt capital structure has less debt than cash plus marketable securities. The baseline sample has 9,866 firms with at least one year of post-peak data on Compustat. Results for the baseline sample are presented in the first set of columns in both panels. Sample sizes for the other subsample entries in both panels are in row 13 of Table 2. Figure 4 View largeDownload slide Median cash-balance ratios and incidence of negative net debt capital structures at the troughs before and after peak market leverage Market leverage (ML) is the ratio of the book value of debt to the sum of the book value of debt and the market value of equity. Peak leverage is the maximum value of ML over a firm’s entire time in the sample. The ML trough after (before) peak is the lowest ML value that comes after (before) the ML peak. Cash/TA is the ratio of cash plus marketable securities to total assets. A negative net debt capital structure has less debt than cash plus marketable securities. The baseline sample has 9,866 firms with at least one year of post-peak data on Compustat. Results for the baseline sample are presented in the first set of columns in both panels. Sample sizes for the other subsample entries in both panels are in row 13 of Table 2. 3. Economic Significance of Decisions that Reduce Leverage The evidence in this section establishes that decisions to repay debt and retain earnings account for a remarkably large portion of observed peak-to-trough declines in ML for our sample firms. Although retention makes an important direct contribution to deleveraging by increasing the denominator of the ML ratio, the resultant impact on ML is typically much smaller than that of debt repayment, and its influence is far from uniform. Most notably, the direct impact of retention on ML is especially important when ML is high and when firms take on more debt while reducing ML. Moreover, sample firms rarely deleverage from peak to near-zero ML simply by retaining earnings. Decisions to issue shares typically make a small direct contribution to reducing ML ratios in the deleveraging episodes we study. However, the dollar amounts of share issuance and retention are both typically large relative to the amount of debt repaid (and to cash balances), which indicates that new equity capital makes a large indirect (funding-related) contribution to deleveraging. The evidence in this section also shows that the rate and scale of deleveraging are often dampened by managerial decisions to accumulate markedly larger cash balances. The fact that managers typically increase cash holdings by material amounts while deleveraging indicates they are concerned with building financial flexibility generally, and are not simply focused on reducing ML. At the same time, the data show that decisions to increase dividends often reduce the size and rate of ML decreases at our sample firms, while also attenuating the cash-balance build up that typically accompanies deleveraging. The latter findings indicate that managerial attempts to rebuild financial flexibility through deleveraging and cash build up are often muted as managers simultaneously take actions to meet the conflicting objective of delivering increased payouts to shareholders. 3.1 Decisions to repay debt, retain earnings, and issue shares Table 3 gauges the size of the contributions of debt repayment, earnings retention, and share issuance decisions to the deleveraging episodes for the sample of 4,476 firms that have five or more years of post-peak data (panel A) and for the baseline sample of 9,866 firms with at least one year of post-peak data (panel B). We focus the discussion on the panel A results because they give a more accurate picture of the nature of complete deleveraging episodes. The reason is that the baseline sample includes many firms with just a few years of data and so the observable deleveraging by these firms is often incomplete. (We analyze attenuated deleveraging in Section 4.) The main difference between panels A and B is that the results in panel B show economically material, but somewhat smaller impacts on deleveraging of managers’ decisions to repay debt, retain earnings, and issue shares (compare rows 9 to 11). Table 3 Managerial actions that reduce leverage: Debt repayment, earnings retention, and share issuance All firms Repay all debt Repay some debt Increase debt A. Sample of firms with at least five years of data after the ML peak 1. Median market leverage (ML) at peak 0.543 0.287 0.612 0.652 2. Median ML at subsequent trough 0.026 0.000 0.078 0.169 3. Median Cash/TA at ML peak 0.050 0.110 0.043 0.040 4. Median Cash/TA at post-peak ML trough 0.132 0.303 0.097 0.052 5. Median Net Debt/TA at ML peak 0.296 0.136 0.342 0.344 6. Median Net Debt/TA at post-peak ML trough $$-$$0.067 $$-$$0.303 0.008 0.173 7. Managerial actions during deleveraging: $$\qquad$$ Median percentage change in debt $$-$$80.2% $$-$$100.0% $$-$$64.0% 61.0% $$\qquad$$ Median [earnings retention $$\div$$ debt plus equity value at peak] 0.243 0.084 0.243 0.790 $$\qquad$$ Median percentage change in shares outstanding 15.2% 11.6% 12.1% 30.5% $$\qquad$$ Percentage of firms with [asset sales $$\div$$ debt] $$>$$ 0.500 27.2% 28.8% 12.3% 12.9% $$\qquad$$ Median percentage change in cash balances 169.5% 169.3% 143.1% 255.2% $$\qquad$$ Percentage of firms that cut dividends 33.2% 17.7% 38.5% 47.4% $$\qquad$$ Percentage of firms with no interim debt increases 32.1% 56.3% 27.3% 0.4% 8. Median peak-to-trough decline in ML $$-$$0.395 $$-$$0.287 $$-$$0.436 $$-$$0.395 9. Median percentage of decline in ML due to $$\qquad$$ Debt repayment (DR) 71.3% 100.0% 48.5% 0.0% $$\qquad$$ DR and earnings retention (ER) 93.7% 100.0% 82.4% 36.6% $$\qquad$$ DR, ER, and net share issuance (SI) 96.5% 100.0% 90.4% 64.4% 10. Percentage of firms for which $$\qquad$$ 100.0% of ML decline is due to DR, ER, and SI 38.3% 100.0% 15.3% 3.2% $$\qquad$$$$\geqslant 50.0\%$$ of ML decline is due to DR, ER, and SI 88.0% 100.0% 89.4% 65.3% $$\qquad$$$$<$$ 10.0% of ML decline is due to DR, ER, and SI 2.8% 0.0% 1.8% 9.8% 11. Adjusted R$$^{2}$$: Percentage of cross-firm variation in $$\quad$$ deleveraging explained by DR, ER, and SI 81% 100% 76% 68% 12. Median ML if Cash/TA increase were used to repay debt 0.000 0.000 0.039 0.147 13. Number of firms 4,476 1,488 2,186 802 14. Percentage of firms in sample 100.0% 33.2% 48.8% 17.9% B. Sample of firms with at least one year of data after the ML peak 1. Median market leverage (ML) at peak 0.491 0.218 0.544 0.579 2. Median ML at subsequent trough 0.088 0.000 0.142 0.242 3. Median Cash/TA at ML peak 0.056 0.150 0.048 0.042 4. Median Cash/TA at post-peak ML trough 0.109 0.309 0.086 0.052 5. Median Net Debt/TA at ML peak 0.283 0.059 0.317 0.333 6. Median Net Debt/TA at post-peak ML trough 0.007 $$-$$0.309 0.064 0.243 7. Managerial actions during deleveraging: $$\qquad$$ Median percentage change in debt $$-$$52.7% $$-$$100.0% $$-$$48.4% 29.1% $$\qquad$$ Median [earnings retention $$\div$$ debt plus equity value at peak] 0.052 0.014 0.049 0.109 $$\qquad$$ Median percentage change in shares outstanding 5.1% 7.5% 3.6% 9.8% $$\qquad$$ Percentage of firms with [asset sales $$\div$$ debt] $$>$$ 0.500 17.6% 26.4% 13.4% 20.1% $$\qquad$$ Median percentage change in cash balances 52.4% 84.6% 36.5% 77.6% $$\qquad$$ Percentage of firms that cut dividends 26.1% 14.8% 28.3% 33.0% $$\qquad$$ Percentage of firms with no interim debt increases 52.2% 68.5% 61.7% 2.2% 8. Median peak-to-trough decline in ML $$-$$0.244 $$-$$0.218 $$-$$0.258 $$-$$0.220 9. Median percentage of decline in ML due to $$\qquad$$ Debt repayment (DR) 57.9% 100.0% 52.8% 0.0% $$\qquad$$ DR and earnings retention (ER) 83.9% 100.0% 78.1% 4.1% $$\qquad$$ DR, ER, and net share issuance (SI) 90.9% 100.0% 86.9% 45.0% 10. Percentage of firms for which $$\qquad$$ 100.0% of ML decline is due to DR, ER, and SI 36.0% 100.0% 23.6% 6.4% $$\qquad$$$$\geqslant 50.0\%$$ of ML decline is due to DR, ER, and SI 78.1% 100.0% 80.9% 45.8% $$\qquad$$$$<$$ 10.0% of ML decline is due to DR, ER, and SI 7.3% 0.0% 3.3% 28.0% 11. Adjusted R$$^{\mathrm{2}}$$: Percentage of cross-firm variation in $$\qquad$$ deleveraging explained by DR, ER, and SI 81% 100% 77% 72% 12. Median ML if Cash/TA increase were used to repay debt 0.054 0.000 0.105 0.222 13. Number of firms 9,866 2,252 5,782 1,832 14. Percentage of firms in sample 100.0% 22.8% 58.6% 18.6% All firms Repay all debt Repay some debt Increase debt A. Sample of firms with at least five years of data after the ML peak 1. Median market leverage (ML) at peak 0.543 0.287 0.612 0.652 2. Median ML at subsequent trough 0.026 0.000 0.078 0.169 3. Median Cash/TA at ML peak 0.050 0.110 0.043 0.040 4. Median Cash/TA at post-peak ML trough 0.132 0.303 0.097 0.052 5. Median Net Debt/TA at ML peak 0.296 0.136 0.342 0.344 6. Median Net Debt/TA at post-peak ML trough $$-$$0.067 $$-$$0.303 0.008 0.173 7. Managerial actions during deleveraging: $$\qquad$$ Median percentage change in debt $$-$$80.2% $$-$$100.0% $$-$$64.0% 61.0% $$\qquad$$ Median [earnings retention $$\div$$ debt plus equity value at peak] 0.243 0.084 0.243 0.790 $$\qquad$$ Median percentage change in shares outstanding 15.2% 11.6% 12.1% 30.5% $$\qquad$$ Percentage of firms with [asset sales $$\div$$ debt] $$>$$ 0.500 27.2% 28.8% 12.3% 12.9% $$\qquad$$ Median percentage change in cash balances 169.5% 169.3% 143.1% 255.2% $$\qquad$$ Percentage of firms that cut dividends 33.2% 17.7% 38.5% 47.4% $$\qquad$$ Percentage of firms with no interim debt increases 32.1% 56.3% 27.3% 0.4% 8. Median peak-to-trough decline in ML $$-$$0.395 $$-$$0.287 $$-$$0.436 $$-$$0.395 9. Median percentage of decline in ML due to $$\qquad$$ Debt repayment (DR) 71.3% 100.0% 48.5% 0.0% $$\qquad$$ DR and earnings retention (ER) 93.7% 100.0% 82.4% 36.6% $$\qquad$$ DR, ER, and net share issuance (SI) 96.5% 100.0% 90.4% 64.4% 10. Percentage of firms for which $$\qquad$$ 100.0% of ML decline is due to DR, ER, and SI 38.3% 100.0% 15.3% 3.2% $$\qquad$$$$\geqslant 50.0\%$$ of ML decline is due to DR, ER, and SI 88.0% 100.0% 89.4% 65.3% $$\qquad$$$$<$$ 10.0% of ML decline is due to DR, ER, and SI 2.8% 0.0% 1.8% 9.8% 11. Adjusted R$$^{2}$$: Percentage of cross-firm variation in $$\quad$$ deleveraging explained by DR, ER, and SI 81% 100% 76% 68% 12. Median ML if Cash/TA increase were used to repay debt 0.000 0.000 0.039 0.147 13. Number of firms 4,476 1,488 2,186 802 14. Percentage of firms in sample 100.0% 33.2% 48.8% 17.9% B. Sample of firms with at least one year of data after the ML peak 1. Median market leverage (ML) at peak 0.491 0.218 0.544 0.579 2. Median ML at subsequent trough 0.088 0.000 0.142 0.242 3. Median Cash/TA at ML peak 0.056 0.150 0.048 0.042 4. Median Cash/TA at post-peak ML trough 0.109 0.309 0.086 0.052 5. Median Net Debt/TA at ML peak 0.283 0.059 0.317 0.333 6. Median Net Debt/TA at post-peak ML trough 0.007 $$-$$0.309 0.064 0.243 7. Managerial actions during deleveraging: $$\qquad$$ Median percentage change in debt $$-$$52.7% $$-$$100.0% $$-$$48.4% 29.1% $$\qquad$$ Median [earnings retention $$\div$$ debt plus equity value at peak] 0.052 0.014 0.049 0.109 $$\qquad$$ Median percentage change in shares outstanding 5.1% 7.5% 3.6% 9.8% $$\qquad$$ Percentage of firms with [asset sales $$\div$$ debt] $$>$$ 0.500 17.6% 26.4% 13.4% 20.1% $$\qquad$$ Median percentage change in cash balances 52.4% 84.6% 36.5% 77.6% $$\qquad$$ Percentage of firms that cut dividends 26.1% 14.8% 28.3% 33.0% $$\qquad$$ Percentage of firms with no interim debt increases 52.2% 68.5% 61.7% 2.2% 8. Median peak-to-trough decline in ML $$-$$0.244 $$-$$0.218 $$-$$0.258 $$-$$0.220 9. Median percentage of decline in ML due to $$\qquad$$ Debt repayment (DR) 57.9% 100.0% 52.8% 0.0% $$\qquad$$ DR and earnings retention (ER) 83.9% 100.0% 78.1% 4.1% $$\qquad$$ DR, ER, and net share issuance (SI) 90.9% 100.0% 86.9% 45.0% 10. Percentage of firms for which $$\qquad$$ 100.0% of ML decline is due to DR, ER, and SI 36.0% 100.0% 23.6% 6.4% $$\qquad$$$$\geqslant 50.0\%$$ of ML decline is due to DR, ER, and SI 78.1% 100.0% 80.9% 45.8% $$\qquad$$$$<$$ 10.0% of ML decline is due to DR, ER, and SI 7.3% 0.0% 3.3% 28.0% 11. Adjusted R$$^{\mathrm{2}}$$: Percentage of cross-firm variation in $$\qquad$$ deleveraging explained by DR, ER, and SI 81% 100% 77% 72% 12. Median ML if Cash/TA increase were used to repay debt 0.054 0.000 0.105 0.222 13. Number of firms 9,866 2,252 5,782 1,832 14. Percentage of firms in sample 100.0% 22.8% 58.6% 18.6% Market leverage (ML) is book debt divided by the sum of book debt and the market value of equity. Peak leverage is the maximum ML over a firm’s time in the sample. The subsequent trough is the lowest value of ML after the peak. Panel A contains 4,476 firms with five or more years of post-peak data on Compustat. Panel B contains the 9,866 firms with at least one year of post-peak data. In row 9 of both panels, we report for the median firm the percentage of the decline in ML (from peak to trough) that would hypothetically result if the only things that changed were the managerial actions specified in the first column; see the appendix for calculation details. In the far-right column, virtually all firms increased debt over the period from peak to trough, while a few firms (0.4% in panel A and 2.2% in panel B) did not decrease debt while deleveraging. Table 3 Managerial actions that reduce leverage: Debt repayment, earnings retention, and share issuance All firms Repay all debt Repay some debt Increase debt A. Sample of firms with at least five years of data after the ML peak 1. Median market leverage (ML) at peak 0.543 0.287 0.612 0.652 2. Median ML at subsequent trough 0.026 0.000 0.078 0.169 3. Median Cash/TA at ML peak 0.050 0.110 0.043 0.040 4. Median Cash/TA at post-peak ML trough 0.132 0.303 0.097 0.052 5. Median Net Debt/TA at ML peak 0.296 0.136 0.342 0.344 6. Median Net Debt/TA at post-peak ML trough $$-$$0.067 $$-$$0.303 0.008 0.173 7. Managerial actions during deleveraging: $$\qquad$$ Median percentage change in debt $$-$$80.2% $$-$$100.0% $$-$$64.0% 61.0% $$\qquad$$ Median [earnings retention $$\div$$ debt plus equity value at peak] 0.243 0.084 0.243 0.790 $$\qquad$$ Median percentage change in shares outstanding 15.2% 11.6% 12.1% 30.5% $$\qquad$$ Percentage of firms with [asset sales $$\div$$ debt] $$>$$ 0.500 27.2% 28.8% 12.3% 12.9% $$\qquad$$ Median percentage change in cash balances 169.5% 169.3% 143.1% 255.2% $$\qquad$$ Percentage of firms that cut dividends 33.2% 17.7% 38.5% 47.4% $$\qquad$$ Percentage of firms with no interim debt increases 32.1% 56.3% 27.3% 0.4% 8. Median peak-to-trough decline in ML $$-$$0.395 $$-$$0.287 $$-$$0.436 $$-$$0.395 9. Median percentage of decline in ML due to $$\qquad$$ Debt repayment (DR) 71.3% 100.0% 48.5% 0.0% $$\qquad$$ DR and earnings retention (ER) 93.7% 100.0% 82.4% 36.6% $$\qquad$$ DR, ER, and net share issuance (SI) 96.5% 100.0% 90.4% 64.4% 10. Percentage of firms for which $$\qquad$$ 100.0% of ML decline is due to DR, ER, and SI 38.3% 100.0% 15.3% 3.2% $$\qquad$$$$\geqslant 50.0\%$$ of ML decline is due to DR, ER, and SI 88.0% 100.0% 89.4% 65.3% $$\qquad$$$$<$$ 10.0% of ML decline is due to DR, ER, and SI 2.8% 0.0% 1.8% 9.8% 11. Adjusted R$$^{2}$$: Percentage of cross-firm variation in $$\quad$$ deleveraging explained by DR, ER, and SI 81% 100% 76% 68% 12. Median ML if Cash/TA increase were used to repay debt 0.000 0.000 0.039 0.147 13. Number of firms 4,476 1,488 2,186 802 14. Percentage of firms in sample 100.0% 33.2% 48.8% 17.9% B. Sample of firms with at least one year of data after the ML peak 1. Median market leverage (ML) at peak 0.491 0.218 0.544 0.579 2. Median ML at subsequent trough 0.088 0.000 0.142 0.242 3. Median Cash/TA at ML peak 0.056 0.150 0.048 0.042 4. Median Cash/TA at post-peak ML trough 0.109 0.309 0.086 0.052 5. Median Net Debt/TA at ML peak 0.283 0.059 0.317 0.333 6. Median Net Debt/TA at post-peak ML trough 0.007 $$-$$0.309 0.064 0.243 7. Managerial actions during deleveraging: $$\qquad$$ Median percentage change in debt $$-$$52.7% $$-$$100.0% $$-$$48.4% 29.1% $$\qquad$$ Median [earnings retention $$\div$$ debt plus equity value at peak] 0.052 0.014 0.049 0.109 $$\qquad$$ Median percentage change in shares outstanding 5.1% 7.5% 3.6% 9.8% $$\qquad$$ Percentage of firms with [asset sales $$\div$$ debt] $$>$$ 0.500 17.6% 26.4% 13.4% 20.1% $$\qquad$$ Median percentage change in cash balances 52.4% 84.6% 36.5% 77.6% $$\qquad$$ Percentage of firms that cut dividends 26.1% 14.8% 28.3% 33.0% $$\qquad$$ Percentage of firms with no interim debt increases 52.2% 68.5% 61.7% 2.2% 8. Median peak-to-trough decline in ML $$-$$0.244 $$-$$0.218 $$-$$0.258 $$-$$0.220 9. Median percentage of decline in ML due to $$\qquad$$ Debt repayment (DR) 57.9% 100.0% 52.8% 0.0% $$\qquad$$ DR and earnings retention (ER) 83.9% 100.0% 78.1% 4.1% $$\qquad$$ DR, ER, and net share issuance (SI) 90.9% 100.0% 86.9% 45.0% 10. Percentage of firms for which $$\qquad$$ 100.0% of ML decline is due to DR, ER, and SI 36.0% 100.0% 23.6% 6.4% $$\qquad$$$$\geqslant 50.0\%$$ of ML decline is due to DR, ER, and SI 78.1% 100.0% 80.9% 45.8% $$\qquad$$$$<$$ 10.0% of ML decline is due to DR, ER, and SI 7.3% 0.0% 3.3% 28.0% 11. Adjusted R$$^{\mathrm{2}}$$: Percentage of cross-firm variation in $$\qquad$$ deleveraging explained by DR, ER, and SI 81% 100% 77% 72% 12. Median ML if Cash/TA increase were used to repay debt 0.054 0.000 0.105 0.222 13. Number of firms 9,866 2,252 5,782 1,832 14. Percentage of firms in sample 100.0% 22.8% 58.6% 18.6% All firms Repay all debt Repay some debt Increase debt A. Sample of firms with at least five years of data after the ML peak 1. Median market leverage (ML) at peak 0.543 0.287 0.612 0.652 2. Median ML at subsequent trough 0.026 0.000 0.078 0.169 3. Median Cash/TA at ML peak 0.050 0.110 0.043 0.040 4. Median Cash/TA at post-peak ML trough 0.132 0.303 0.097 0.052 5. Median Net Debt/TA at ML peak 0.296 0.136 0.342 0.344 6. Median Net Debt/TA at post-peak ML trough $$-$$0.067 $$-$$0.303 0.008 0.173 7. Managerial actions during deleveraging: $$\qquad$$ Median percentage change in debt $$-$$80.2% $$-$$100.0% $$-$$64.0% 61.0% $$\qquad$$ Median [earnings retention $$\div$$ debt plus equity value at peak] 0.243 0.084 0.243 0.790 $$\qquad$$ Median percentage change in shares outstanding 15.2% 11.6% 12.1% 30.5% $$\qquad$$ Percentage of firms with [asset sales $$\div$$ debt] $$>$$ 0.500 27.2% 28.8% 12.3% 12.9% $$\qquad$$ Median percentage change in cash balances 169.5% 169.3% 143.1% 255.2% $$\qquad$$ Percentage of firms that cut dividends 33.2% 17.7% 38.5% 47.4% $$\qquad$$ Percentage of firms with no interim debt increases 32.1% 56.3% 27.3% 0.4% 8. Median peak-to-trough decline in ML $$-$$0.395 $$-$$0.287 $$-$$0.436 $$-$$0.395 9. Median percentage of decline in ML due to $$\qquad$$ Debt repayment (DR) 71.3% 100.0% 48.5% 0.0% $$\qquad$$ DR and earnings retention (ER) 93.7% 100.0% 82.4% 36.6% $$\qquad$$ DR, ER, and net share issuance (SI) 96.5% 100.0% 90.4% 64.4% 10. Percentage of firms for which $$\qquad$$ 100.0% of ML decline is due to DR, ER, and SI 38.3% 100.0% 15.3% 3.2% $$\qquad$$$$\geqslant 50.0\%$$ of ML decline is due to DR, ER, and SI 88.0% 100.0% 89.4% 65.3% $$\qquad$$$$<$$ 10.0% of ML decline is due to DR, ER, and SI 2.8% 0.0% 1.8% 9.8% 11. Adjusted R$$^{2}$$: Percentage of cross-firm variation in $$\quad$$ deleveraging explained by DR, ER, and SI 81% 100% 76% 68% 12. Median ML if Cash/TA increase were used to repay debt 0.000 0.000 0.039 0.147 13. Number of firms 4,476 1,488 2,186 802 14. Percentage of firms in sample 100.0% 33.2% 48.8% 17.9% B. Sample of firms with at least one year of data after the ML peak 1. Median market leverage (ML) at peak 0.491 0.218 0.544 0.579 2. Median ML at subsequent trough 0.088 0.000 0.142 0.242 3. Median Cash/TA at ML peak 0.056 0.150 0.048 0.042 4. Median Cash/TA at post-peak ML trough 0.109 0.309 0.086 0.052 5. Median Net Debt/TA at ML peak 0.283 0.059 0.317 0.333 6. Median Net Debt/TA at post-peak ML trough 0.007 $$-$$0.309 0.064 0.243 7. Managerial actions during deleveraging: $$\qquad$$ Median percentage change in debt $$-$$52.7% $$-$$100.0% $$-$$48.4% 29.1% $$\qquad$$ Median [earnings retention $$\div$$ debt plus equity value at peak] 0.052 0.014 0.049 0.109 $$\qquad$$ Median percentage change in shares outstanding 5.1% 7.5% 3.6% 9.8% $$\qquad$$ Percentage of firms with [asset sales $$\div$$ debt] $$>$$ 0.500 17.6% 26.4% 13.4% 20.1% $$\qquad$$ Median percentage change in cash balances 52.4% 84.6% 36.5% 77.6% $$\qquad$$ Percentage of firms that cut dividends 26.1% 14.8% 28.3% 33.0% $$\qquad$$ Percentage of firms with no interim debt increases 52.2% 68.5% 61.7% 2.2% 8. Median peak-to-trough decline in ML $$-$$0.244 $$-$$0.218 $$-$$0.258 $$-$$0.220 9. Median percentage of decline in ML due to $$\qquad$$ Debt repayment (DR) 57.9% 100.0% 52.8% 0.0% $$\qquad$$ DR and earnings retention (ER) 83.9% 100.0% 78.1% 4.1% $$\qquad$$ DR, ER, and net share issuance (SI) 90.9% 100.0% 86.9% 45.0% 10. Percentage of firms for which $$\qquad$$ 100.0% of ML decline is due to DR, ER, and SI 36.0% 100.0% 23.6% 6.4% $$\qquad$$$$\geqslant 50.0\%$$ of ML decline is due to DR, ER, and SI 78.1% 100.0% 80.9% 45.8% $$\qquad$$$$<$$ 10.0% of ML decline is due to DR, ER, and SI 7.3% 0.0% 3.3% 28.0% 11. Adjusted R$$^{\mathrm{2}}$$: Percentage of cross-firm variation in $$\qquad$$ deleveraging explained by DR, ER, and SI 81% 100% 77% 72% 12. Median ML if Cash/TA increase were used to repay debt 0.054 0.000 0.105 0.222 13. Number of firms 9,866 2,252 5,782 1,832 14. Percentage of firms in sample 100.0% 22.8% 58.6% 18.6% Market leverage (ML) is book debt divided by the sum of book debt and the market value of equity. Peak leverage is the maximum ML over a firm’s time in the sample. The subsequent trough is the lowest value of ML after the peak. Panel A contains 4,476 firms with five or more years of post-peak data on Compustat. Panel B contains the 9,866 firms with at least one year of post-peak data. In row 9 of both panels, we report for the median firm the percentage of the decline in ML (from peak to trough) that would hypothetically result if the only things that changed were the managerial actions specified in the first column; see the appendix for calculation details. In the far-right column, virtually all firms increased debt over the period from peak to trough, while a few firms (0.4% in panel A and 2.2% in panel B) did not decrease debt while deleveraging. Section 6.3 reports robustness checks that indicate that our main conclusions are unchanged when we analyze deleveraging after large leverage increases rather than, as in Table 3, after firms attain their all-time peak ML ratios. Our main inferences are also the same when we analyze deleveraging from peak leverage measured in book value terms; see Table IA1 in the Internet Appendix. Table 3 reports findings for all 4,476 sample firms and for subsamples of firms that, over the period from ML peak to trough, repay all debt (1,488 firms); repay some, but not all, debt (2,186 firms); and increase debt, or in a few cases, leave it unchanged (802 firms). Firms in the first group obviously end their deleveraging episodes with ML $$=$$ 0.000, while those in the latter two groups deleverage to low but positive ML ratios (row 2, panel A, Table 3). For firms in the third group, increases in the total market value of equity outweigh the increase in debt, and so their ML ratios decline. Table 3 reveals considerable heterogeneity across the three groups in the magnitude of their ML reductions (rows 1, 2, and 8; panel A). Among firms that repay all debt, median ML declines from 0.287 at the peak to 0.000 at the later trough. The median ML ratio declines by larger amounts at firms that repay some, but not all, debt and at firms that increase debt. The former firms show an analogous decrease of 0.534 (i.e., 0.612 at the peak minus 0.078 at the trough), while the latter firms show a decrease of 0.483 (i.e., 0.652 minus 0.169). The median peak-to-trough change in ML is somewhat smaller than the change in the median ratio, but it is nonetheless substantial for all debt categories (compare the differences between rows 1 and 2 with the corresponding entry in row 8, Table 3). The ML figures in Table 3 materially understate the size of deleveraging for firms that repay all debt. The reason is that ML ratios are bounded below at 0.000, and this limit is binding when firms repay all debt. Firms that pay off all debt can nevertheless continue to deleverage simply by accumulating larger cash holdings. They do so with a vengeance, with the median firm almost tripling Cash/TA from 0.110 to 0.303 as ML moves from peak to trough (rows 3 and 4, panel A). The important point is that a nontrivial subset of firms repays all debt and accumulates large cash balances. In so doing, these firms drive their net-debt ratios deep into negative territory. This is not random or passive down-drifting of ML and net-debt ratios. The reason is that paying down debt and building cash balances are decisions of the firm, not exogenous shocks. Table 3 further shows that the median firm in the full sample repays 80.2% of the debt outstanding at peak ML, retains additional earnings equal to 24.3% of the total value of debt and equity at peak ML, and increases the number of outstanding shares by 15.2% (row 7, panel A). We focus here on debt repayment, earnings retention, and share issuance because each of these managerial decisions directly affects the ML ratio. Debt repayment reduces both the numerator and denominator of the ratio, while only the denominator increases directly with additional equity. While internally generated equity and stock-sale proceeds both provide cash that could be used for debt repayment, they both reduce ML even when debt is not repaid. Table 3 takes into account only the direct impact of equity expansions on the denominator of the ML ratio. Debt repayment is taken into account regardless of the source of the funds for the repayment. Section 3.6 documents that, among sample firms that repay debt, share issuance and retention both provide an amount of funds that loom large relative to the size of debt repayment and cash balances at the end of the deleveraging episode. Many sample firms sell substantial amounts of assets (row 7, panel A, Table 3), but their receipt of asset-sale proceeds does not directly affect any element of their ML ratios. Asset-sale proceeds provide the firm with cash that could be, but is not necessarily, used to repay debt. Suppose the firm does not repay debt and instead invests the proceeds elsewhere. Leaving aside any value change from the asset disposal itself or from the new investment, the firm’s total (debt plus equity) value is unchanged, as is its debt. Hence, its ML ratio is unchanged by the asset sale. We accordingly do not treat asset-sale proceeds as an element of proactive managerial deleveraging that is distinct from debt repayment. Rather, we treat asset sales simply as a source of cash that makes it possible for firms to repay debt. Section 3.6 reports that, among firms that repay debt, asset sale proceeds are a nontrivial source of funds, albeit markedly smaller than resources provided through share issuance and earnings retention. The fact that many sample firms undertake large asset sales around the time of peak ML is, of course, suggestive of material proactive deleveraging, with firms selling assets because managers intend to use the proceeds to repay debt. See the appendix in Lang, Poulsen, and Stulz (1995) for evidence of pervasive use of asset sales to fund debt repayment. Quite a few sample firms also free up cash by reducing dividends during their deleveraging episodes. Consistent with Section 4’s evidence of a nontrivial incidence of financial distress at sample firms, we find that 33.2% of firms cut dividends at some point while deleveraging (row 7, panel A, Table 3). Since dividend cuts are reflected in the net amount of retained earnings, we simply treat them as a source of cash and not as an additional element of proactive deleveraging by management. The important bottom line from row 7 of Table 3 is that, because the magnitudes of debt repayments, earnings retention, and share issuance are nontrivial, these managerial decisions could plausibly be responsible for a significant portion of sample firms’ large-scale deleveraging. 3.2 Portion of observed deleveraging due to managerial decisions To measure the contribution of these managerial actions to the scale of deleveraging, we adopt an approach similar in spirit to Welch’s (2004) analysis of the extent to which actual variation in ML is due to security issuance and retirement decisions. We first assess for each firm what ML would be at the trough after peak if the specific actions in question are the only things that change from the time the firm is at peak ML. We use the acronym HML to distinguish hypothetical from actual ML values. The portion of the firm’s actual deleveraging explained by the action(s) in question is then the percentage equivalent of [ML(peak) – HML(trough)] $$\div$$ [ML(peak) – ML(trough)]. This ratio is bounded by 0.0% and 100.0% by the algorithm that generates values of HML(trough), which is described in the appendix. Two of the assumptions of the algorithm are sufficiently important to merit highlighting here. First, the algorithm assumes that earnings retention translates dollar-for-dollar into a higher market value of equity. This assumption means that we ignore agency costs and taxes and make no upward adjustment to equity value for future earnings (above the appropriate risk-adjusted cost of capital) on resources that are retained after peak but before the firm reaches its post-peak ML trough. Second, the net share-issuance effect is based on the open-market share price near the time of each issuance and is therefore an estimate of the impact on the ML ratio from issuing shares during the time between peak and trough. Table 3 indicates that decisions to repay debt, retain earnings, and issue shares together account for 96.5% of the actual deleveraging by the median firm in the full sample (row 8). Debt repayment alone accounts for 71.3% of the median firm’s deleveraging, while debt repayment and earnings retention account for 93.7%. A comparison of the 71.3% and 93.7% figures indicates that, as with decisions to repay debt, decisions to retain earnings account for a substantial portion of the typical deleveraging episode in our sample.3 A comparison of the 93.7% and 96.5% figures indicates that the direct impact of net share issuance typically accounts for only a small increment to deleveraging above debt repayment and retention. However, as discussed further below, the indirect impact of share issuance, namely its impact on the funds available to pay down debt, is economically much more substantial than the direct impact on deleveraging. Decisions to repay debt, retain earnings, and issue shares collectively account for 100.0% of the observed deleveraging for 38.0% of sample firms (row 9, panel A, Table 3). The three decisions together account for at least 50.0% of the deleveraging for 88.0% of firms, and they are responsible for less than 10.0% of the actual deleveraging for only 2.8% of firms (row 10). These decisions also account for a large portion of cross-firm variation in deleveraging. The adjusted R$$^{2}$$ is 81% for a regression in which the dependent variable is the actual deleveraging from peak to trough and the right-hand side variables are a constant and the hypothetical deleveraging over the same period due solely to debt repayment, earnings retention, and share issuance (row 11, Table 3). Although debt repayment accounts for 71.3% of the deleveraging at the median sample firm, it plays no role whatsoever for the 17.9% of firms that increase debt while deleveraging (row 9, Table 3). If the ML ratio decreases despite an increase in the dollar amount of debt, then obviously total equity value must have increased. What is not obvious is whether total equity value increased because managers took actions to increase equity capital and, if managers in fact took such actions, how much of the resultant ML decrease reflects decisions to increase internally generated equity (through earnings retention) as opposed to increase externally generated equity (through new share issuance). Among firms that increase debt while deleveraging, earnings retention accounts for 36.6% of the median firm’s peak-to-trough decline in ML, while retention plus share issuance together account for 64.4% (row 9, Table 3). Thus, decisions to increase equity capital account for over half the total decline in ML for the median firm in this subsample, with both internal equity (generated by earnings retention) and external equity (obtained from share issuance) contributing significantly to the typical deleveraging. Table 3 treats the post-peak trough ML as the deleveraging outcome, yet doing so may overstate the extent to which the outcome is due to managers’ decisions. The reason is that positive shocks may have raised equity values significantly in the trough year so that a decline to a near-zero ML in that year could be largely exogenous for many firms (even though the full peak-to-trough decline is mostly endogenous). The data strongly reject this conjecture, indicating instead that the last annual step down to trough ML is, for most sample firms, due almost entirely to decisions made by managers. We reach this conclusion using the same approach as in Table 3, but now gauging the proactive portion of the trough-year decline in ML, with details reported next, but not tabulated. For the median firm in the baseline sample, the trough-year decline in ML is 36.0% of the full decline from peak to trough and 89.2% of the trough-year decline is due to decisions to repay debt, retain earnings, and/or issue stock. For firms with at least five years of post-peak data, the analogous figures are 13.7% and 96.9%. Debt repayment, retention, and share issuance together account for more than half of the trough-year decline in ML for 69.4% of firms in the baseline sample and for 74.3% of firms with at least five years of post-peak data. 3.3 Cash-balance build-up: Restoring financial flexibility generally versus simply deleveraging Table 3 further reveals that large increases in cash balances typically accompany the deleveraging episodes we study (rows 3, 4, and 7). This cash accumulation increases financial flexibility in terms of providing more resources inside the firm that can be used immediately. In this sense, a cash buildup is a complement to an ML reduction, which gives a firm greater unused debt capacity than it would otherwise have. However, cash accumulation and debt repayment decisions are also substitutes because excess cash holdings could be used to pay down debt. If cash accumulated during deleveraging were used to pay down debt, the median firm’s ML at the trough after peak would have been 0.000 rather than 0.026 (row 12). Thus, the typical sample firm could have had a zero-debt capital structure had it not accumulated as much in cash balances and instead paid down more debt. These findings indicate that most sample firms are concerned with rebuilding financial flexibility generally rather than with simply seeking to reduce ML. This interpretation reflects the fact that cash balances and unused debt capacity are imperfect substitute sources of flexibility, with a given amount of cash balances actually providing greater assured access to capital than the same amount of unused debt capacity. At the same time, however, stockpiling cash will not always dominate paying down debt (replenishing debt capacity). Disincentives for cash stockpiling come from corporate taxes, agency costs, and/or a market premium (an interest-rate discount) on liquid asset holdings attributable to the advantage that cash balances have as a reliable source of capital. 3.4 Earnings retention and endogenous deleveraging Most prior empirical studies ignore the fact that decisions to retain earnings affect ML ratios through their impact on the total market value of equity. Instead, the literature’s usual approach has been to treat security issuance and retirement decisions as the sole decision variables that alter leverage, with all non-issuance-related equity-value changes interpreted as exogenous shocks to ML. For example, Welch’s (2004) influential study follows the usual approach when decomposing ML changes into endogenous and exogenous components, thereby implicitly treating retention as an exogenous shock that reduces ML. To see why we view retention as a managerial decision, the key is to distinguish between earnings and retention of earnings. For a given firm, earnings arrive each period and some portion of an earnings realization may include an unanticipated exogenous component. However, whatever the amount of any positive earnings realization, the amount of earnings kept in the firm is a choice. The “earnings-retention decision” refers to the choice about how much of currently earned resources to keep versus how much to distribute to shareholders. The earnings-retention decision is just different terminology to describe the decision of how much of earnings to pay out; see, for example, Brennan (1971, p. 1116), who parameterizes a firm’s choice of dividend policy by its choice of retention ratio. For our study, the important point is that the level of earnings retention is a decision made by managers that determines the amount of internally generated capital that is added to the equity portion of a firm’s debt-equity mix. We would also note that a failure to treat earnings retention as endogenous involves an arbitrary and unwarranted asymmetry in the treatment of security issuance/retirement and retention decisions. If a firm repurchases equity, the literature rightly classifies this security retirement action as a decision to increase ML. Symmetrically, a decision not to repurchase equity and, instead, to retain earnings is a decision to decrease ML. Payouts raise ML. Nonpayouts (retention) reduce ML. There is accordingly an inherent logical inconsistency in treating stock repurchase (payout) decisions as a vehicle through which managers alter a firm’s ML ratio while, at the same time, ignoring the ML impact of retention. Earnings retention is endogenous to managers even if dividends are “sticky” in the sense of Lintner (1956). With sticky dividends, a firm’s payouts tend to grow smoothly over time, even when there is a sharp change in earnings.4 In this case, although some portion of an increase in earnings will tend to be retained in the firm, the resultant retention is the consequence of managers choosing how much of each earnings realization to keep or distribute. For example, with each increase in earnings, managers could declare a special dividend or repurchase shares and alter to any degree they want (including elimination) the incremental earnings retention. Retained earnings therefore do not simply pile up inside a firm due to constraints that put the level of retention outside the control of managers. In short, the amount of earnings retention is chosen, hence endogenous. 3.5 Heterogeneity in the deleveraging impact of earnings retention Table 3 indicates that debt repayment accounts for 71.3% of the peak-to-trough decline in ML for the median firm that has at least five years of post-peak data, and that earnings retention and debt repayment together account for 93.7% (row 9, first column, panel A). These findings indicate that, while debt repayment is the main proactive component of deleveraging, earnings retention also makes a nontrivial direct contribution to the deleveraging of the typical sample firm. A closer look at the data reveals substantial cross-firm heterogeneity in the impact of retention in the peak-to-trough decline in ML. Specifically, in Table 3, the importance of earnings retention stands out most clearly for the 17.9% of firms that reduce ML while simultaneously increasing debt. During its deleveraging episode, the median firm in this subsample adds retained earnings equal to 79.0% of the total value of debt plus equity that prevailed at peak ML (row 7, last columns, panel A). This incremental earnings retention translates to a 79.0% increase in the denominator of the ML ratio which, other things equal, cuts the median firm’s ML ratio almost in half from the 0.652 ratio that prevailed at peak. Although Table 3 shows that the direct deleveraging impact of retention is economically material in our sample, its impact is largely felt when firms have high to moderate ML ratios, and it is rare to find that retention alone drives a firm to a low leverage capital structure. Among the 9,866 firms in our baseline sample, 5,121 have ML below 0.100 at the post-peak trough. Only 0.7% (38 of these firms) deleveraged to an ML ratio below 0.100 with positive earnings retention (between peak and trough) and with no debt pay down and no share issuance (not tabulated). The small direct deleveraging impact of earnings retention at low levels of ML reflects the fact that leverage is measured as the fraction of total value that comes from debt. Because retention increases equity value in the denominator of the ML ratio, the impact of any increase in equity value (including retention) on ML is nonlinear and declines monotonically as the beginning level of ML declines. A given large amount of retention generates a large reduction in ML when ML is initially high, with the ML impact of the equity increase becoming smaller at lower initial levels of ML. For example, it would take earnings retention equal to (1) 100% of total firm value to reduce ML from 0.500 to 0.250 and (2) 900% of firm value to reduce ML from 0.500 to 0.050. Thus, it would take an enormous amount of retention (relative to firm value) to depress the ML ratio from its typical level at peak (around 0.500) to a near-zero level. 3.6 Indirect contributions to deleveraging: Funds for debt repayment The direct deleveraging effects of earnings retention and share issuance reported in Table 3 could materially understate the importance of new equity capital for deleveraging and, more generally, for bolstering financial flexibility. The reason is that both forms of new equity provide resources that could be used to repay debt (thereby bolstering flexibility in the form of unused debt capacity by reducing ML ratios) or to increase cash balances (thereby bolstering flexibility in the form of spendable funds that are immediately available to managers).5 We gauge this possibility in Table 4, which restricts attention to firms that repaid some or all of the debt that was outstanding at peak ML (see the second and third columns of Table 3) and which considers share-issuance proceeds, earnings retention, and asset-sale proceeds as potential sources of funds. The first three columns report (for the median firm) the magnitude of each of these items relative to the dollar amounts of (1) debt that was outstanding at peak, (2) debt actually repaid, and (3) cash held at the trough after peak ML. The fourth and fifth columns report the median firm’s actual Cash/TA ratio and the hypothetical Cash/TA ratio it would have had absent the specific funds for the row in question. The sixth column reports the percentage of firms that would have been able to cover their debt repayment with the specified source of funds. The last column reports the percentage of firms that, without the particular source of funds and holding everything else constant, would have had zero cash balances at the post-peak trough. Table 4 Share-issuance proceeds, retained earnings, and asset sale proceeds as a percentage of debt repayment and cash balances after deleveraging Value for median firm of specified item as a percentage of Median firm’s cash/total assets ratio at trough Percentage of firms with specified item $$\geqslant$$ Debt at peak Debt repaid Cash at trough Actual Absent row item Debt repaid Cash at trough A. Firms with at least five years of data after the ML peak Share issuance 24.9% 40.1% 41.5% 0.167 0.054 36.5% 33.8% Retained earnings 44.8% 72.7% 78.0% 0.165 0.018 45.7% 45.5% Asset sales 8.2% 13.1% 15.3% 0.166 0.083 22.1% 27.2% Issuance $$+$$ retention 120.3% 189.4% 186.1% 0.167 0.000 64.8% 68.2% Issuance $$+$$ retention $$+$$ asset sales 159.9% 257.0% 264.6% 0.167 0.000 72.6% 77.2% B. Firms with at least one year of data after the ML peak Share issuance 6.8% 14.6% 14.2% 0.131 0.054 29.6% 26.9% Retained earnings 10.9% 28.1% 27.1% 0.130 0.040 34.9% 34.5% Asset sales 2.2% 4.4% 5.4% 0.132 0.072 17.0% 22.1% Issuance $$+$$ retention 52.7% 117.5% 119.7% 0.130 0.000 53.0% 54.3% Issuance $$+$$ retention $$+$$ asset sales 77.0% 161.4% 180.8% 0.131 0.000 60.9% 64.7% Value for median firm of specified item as a percentage of Median firm’s cash/total assets ratio at trough Percentage of firms with specified item $$\geqslant$$ Debt at peak Debt repaid Cash at trough Actual Absent row item Debt repaid Cash at trough A. Firms with at least five years of data after the ML peak Share issuance 24.9% 40.1% 41.5% 0.167 0.054 36.5% 33.8% Retained earnings 44.8% 72.7% 78.0% 0.165 0.018 45.7% 45.5% Asset sales 8.2% 13.1% 15.3% 0.166 0.083 22.1% 27.2% Issuance $$+$$ retention 120.3% 189.4% 186.1% 0.167 0.000 64.8% 68.2% Issuance $$+$$ retention $$+$$ asset sales 159.9% 257.0% 264.6% 0.167 0.000 72.6% 77.2% B. Firms with at least one year of data after the ML peak Share issuance 6.8% 14.6% 14.2% 0.131 0.054 29.6% 26.9% Retained earnings 10.9% 28.1% 27.1% 0.130 0.040 34.9% 34.5% Asset sales 2.2% 4.4% 5.4% 0.132 0.072 17.0% 22.1% Issuance $$+$$ retention 52.7% 117.5% 119.7% 0.130 0.000 53.0% 54.3% Issuance $$+$$ retention $$+$$ asset sales 77.0% 161.4% 180.8% 0.131 0.000 60.9% 64.7% Market leverage (ML) is book debt divided by the sum of book debt and the market value of equity. Peak leverage is the maximum ML over a firm’s time in the sample. The subsequent trough is the lowest value of ML after the peak. The sample here is restricted to firms that repaid some or all debt during their deleveraging episodes. These are the firms in the second and third columns of numbers in Table 3. Panel A is based on the sample of 3,674 firms with five or more years of post-peak data on Compustat, and panel B is based on the 8,034 firms with at least one year of post-peak data. Missing values of some items reduce the sample sizes below 3,674 for panel A and below 8,034 for panel B. In all cases, the calculations are based on at least 82% of the total number of firms in each column, and usually they are based on more than 90% of the total number of firms. The first three columns of numbers refer to, respectively, the dollar value of debt at the time the firm was at peak ML, the dollar value of debt repaid over the entire deleveraging episode, and the dollar value of cash when ML is at the post-peak trough. The fifth column reports (for the median firm) what the Cash/TA ratio would have been absent the funding source specified in the row in question. (TA denotes the total book value of assets.) For example, the entry in the fifth column for the retained earnings (RE) row reports what the Cash/TA ratio would have been for the median firm if the firm had paid out cash on hand in the amount RE. The share issuance amount is based on items reported in the Statement of Cash Flows on Compustat and equals common plus preferred stock issuance proceeds minus the change in the value of preferred stock. Table 4 Share-issuance proceeds, retained earnings, and asset sale proceeds as a percentage of debt repayment and cash balances after deleveraging Value for median firm of specified item as a percentage of Median firm’s cash/total assets ratio at trough Percentage of firms with specified item $$\geqslant$$ Debt at peak Debt repaid Cash at trough Actual Absent row item Debt repaid Cash at trough A. Firms with at least five years of data after the ML peak Share issuance 24.9% 40.1% 41.5% 0.167 0.054 36.5% 33.8% Retained earnings 44.8% 72.7% 78.0% 0.165 0.018 45.7% 45.5% Asset sales 8.2% 13.1% 15.3% 0.166 0.083 22.1% 27.2% Issuance $$+$$ retention 120.3% 189.4% 186.1% 0.167 0.000 64.8% 68.2% Issuance $$+$$ retention $$+$$ asset sales 159.9% 257.0% 264.6% 0.167 0.000 72.6% 77.2% B. Firms with at least one year of data after the ML peak Share issuance 6.8% 14.6% 14.2% 0.131 0.054 29.6% 26.9% Retained earnings 10.9% 28.1% 27.1% 0.130 0.040 34.9% 34.5% Asset sales 2.2% 4.4% 5.4% 0.132 0.072 17.0% 22.1% Issuance $$+$$ retention 52.7% 117.5% 119.7% 0.130 0.000 53.0% 54.3% Issuance $$+$$ retention $$+$$ asset sales 77.0% 161.4% 180.8% 0.131 0.000 60.9% 64.7% Value for median firm of specified item as a percentage of Median firm’s cash/total assets ratio at trough Percentage of firms with specified item $$\geqslant$$ Debt at peak Debt repaid Cash at trough Actual Absent row item Debt repaid Cash at trough A. Firms with at least five years of data after the ML peak Share issuance 24.9% 40.1% 41.5% 0.167 0.054 36.5% 33.8% Retained earnings 44.8% 72.7% 78.0% 0.165 0.018 45.7% 45.5% Asset sales 8.2% 13.1% 15.3% 0.166 0.083 22.1% 27.2% Issuance $$+$$ retention 120.3% 189.4% 186.1% 0.167 0.000 64.8% 68.2% Issuance $$+$$ retention $$+$$ asset sales 159.9% 257.0% 264.6% 0.167 0.000 72.6% 77.2% B. Firms with at least one year of data after the ML peak Share issuance 6.8% 14.6% 14.2% 0.131 0.054 29.6% 26.9% Retained earnings 10.9% 28.1% 27.1% 0.130 0.040 34.9% 34.5% Asset sales 2.2% 4.4% 5.4% 0.132 0.072 17.0% 22.1% Issuance $$+$$ retention 52.7% 117.5% 119.7% 0.130 0.000 53.0% 54.3% Issuance $$+$$ retention $$+$$ asset sales 77.0% 161.4% 180.8% 0.131 0.000 60.9% 64.7% Market leverage (ML) is book debt divided by the sum of book debt and the market value of equity. Peak leverage is the maximum ML over a firm’s time in the sample. The subsequent trough is the lowest value of ML after the peak. The sample here is restricted to firms that repaid some or all debt during their deleveraging episodes. These are the firms in the second and third columns of numbers in Table 3. Panel A is based on the sample of 3,674 firms with five or more years of post-peak data on Compustat, and panel B is based on the 8,034 firms with at least one year of post-peak data. Missing values of some items reduce the sample sizes below 3,674 for panel A and below 8,034 for panel B. In all cases, the calculations are based on at least 82% of the total number of firms in each column, and usually they are based on more than 90% of the total number of firms. The first three columns of numbers refer to, respectively, the dollar value of debt at the time the firm was at peak ML, the dollar value of debt repaid over the entire deleveraging episode, and the dollar value of cash when ML is at the post-peak trough. The fifth column reports (for the median firm) what the Cash/TA ratio would have been absent the funding source specified in the row in question. (TA denotes the total book value of assets.) For example, the entry in the fifth column for the retained earnings (RE) row reports what the Cash/TA ratio would have been for the median firm if the firm had paid out cash on hand in the amount RE. The share issuance amount is based on items reported in the Statement of Cash Flows on Compustat and equals common plus preferred stock issuance proceeds minus the change in the value of preferred stock. Table 4 indicates that share issuance and retained earnings are, individually and jointly, substantial sources of funds for the typical sample firm that repays debt. Panel A shows that, among firms with at least five years of post-peak data, share issuance and earnings retention individually constitute 40.1% and 72.7% of debt repaid by the median firm and, taken together, they constitute 189.4% (first, second, and fourth rows; second column). They also loom large relative to cash balances at the post-peak trough and, by implication, relative to any cash increases during the deleveraging episode (see the same rows; third column). Over two thirds of sample firms (68.2%) would have had no cash if retained earnings had been paid out and no new shares sold (sixth column) and almost as many (64.8%) could fully cover their debt repayment with their new equity (fifth column). Panel B of Table 4 shows that similar, but somewhat more muted, effects are observed for firms with at least one year of post-peak data. The key message from Table 4 is that share-issuance proceeds and retained earnings are economically material indirect contributors to deleveraging in that both are typically large relative to the amounts of debt repaid and to cash holdings at the post-peak trough. Table 4 also shows that asset-sale proceeds are typically nontrivial relative to debt repaid and cash balances, but are generally much smaller than share-issuance proceeds and retention as a percentage of debt repayment and of cash holdings. 3.7 Forgone deleveraging and financial flexibility: Increases in dividends and debt Decisions to increase equity payouts rather than retain earnings (and possibly repay debt with the retained cash) will, of course, mute the extent that firms rebuild financial flexibility. The dampening impact of equity payouts on deleveraging is clear in Denis and McKeon (2012, Figure 4) and it is also apparent in our sample, as can be seen in Table 5, which analyzes samples of firms with at least five years of post-peak data. The first column examines the 38.2% of these firms that increase dividends while deleveraging. The second column examines the 17.9% of these firms that increase their debt obligations while reducing their ML ratios; this sample is the same as that underlying the far-right column in Table 3. Table 5 Forgone deleveraging Actions taken while deleveraging: Increase dividends Increase debt 1. Median market leverage (ML) at $$\qquad$$ peak 0.582 0.652 $$\qquad$$ trough 0.082 0.169 $$\qquad$$ trough if debt repaid and cash stockpiled with dividend increases 0.008 0.056 2. Median Cash/TA ratio at $$\qquad$$ peak ML 0.042 0.040 $$\qquad$$ trough ML 0.089 0.052 $$\qquad$$ trough if debt repaid and cash stockpiled with dividend increases 0.120 0.073 3. Percentage of firms with negative net debt at $$\qquad$$ trough ML 47.0% 18.5% $$\qquad$$ trough if debt repaid and cash stockpiled with dividend increases 64.3% 46.3% 4. Median [earnings retention $$\div$$ debt plus equity value at peak] 0.606 0.790 5. Median dividends $$\div$$ dividends plus earnings retention 0.251 0.272 6. Median asset growth during deleveraging 79.9% 190.3% 7. Percentage of firms that increase dividends while deleveraging 100.0% 66.1% 8. Number of firms 1,711 802 9. Percentage of all firms with 5 or more years of post-peak data 38.2% 17.9% Actions taken while deleveraging: Increase dividends Increase debt 1. Median market leverage (ML) at $$\qquad$$ peak 0.582 0.652 $$\qquad$$ trough 0.082 0.169 $$\qquad$$ trough if debt repaid and cash stockpiled with dividend increases 0.008 0.056 2. Median Cash/TA ratio at $$\qquad$$ peak ML 0.042 0.040 $$\qquad$$ trough ML 0.089 0.052 $$\qquad$$ trough if debt repaid and cash stockpiled with dividend increases 0.120 0.073 3. Percentage of firms with negative net debt at $$\qquad$$ trough ML 47.0% 18.5% $$\qquad$$ trough if debt repaid and cash stockpiled with dividend increases 64.3% 46.3% 4. Median [earnings retention $$\div$$ debt plus equity value at peak] 0.606 0.790 5. Median dividends $$\div$$ dividends plus earnings retention 0.251 0.272 6. Median asset growth during deleveraging 79.9% 190.3% 7. Percentage of firms that increase dividends while deleveraging 100.0% 66.1% 8. Number of firms 1,711 802 9. Percentage of all firms with 5 or more years of post-peak data 38.2% 17.9% The samples of dividend-increasing and debt-increasing firms in this table are from the 4,476 firms in the baseline sample with 5 or more years of post-peak data on Compustat. Dividend-increasing firms are those whose average annual dividend payment from the year after peak to the trough year is larger than the dividend payment in the peak year. The last item in rows 1, 2, and 3 reports the hypothetical value of the variable in question that would have obtained at the trough if (1) the firm first used all of the cash that had been allocated to dividend increases to repay debt and (2) if any of the latter cash were left over after paying debt down to zero, the remainder would be placed in cash balances. The variable in row 5 restricts attention to firms with positive retained earnings during the deleveraging episode. The sample in the far-right column of this table is the same as the sample analyzed in the far-right column of Panel A of Table 3. Table 5 Forgone deleveraging Actions taken while deleveraging: Increase dividends Increase debt 1. Median market leverage (ML) at $$\qquad$$ peak 0.582 0.652 $$\qquad$$ trough 0.082 0.169 $$\qquad$$ trough if debt repaid and cash stockpiled with dividend increases 0.008 0.056 2. Median Cash/TA ratio at $$\qquad$$ peak ML 0.042 0.040 $$\qquad$$ trough ML 0.089 0.052 $$\qquad$$ trough if debt repaid and cash stockpiled with dividend increases 0.120 0.073 3. Percentage of firms with negative net debt at $$\qquad$$ trough ML 47.0% 18.5% $$\qquad$$ trough if debt repaid and cash stockpiled with dividend increases 64.3% 46.3% 4. Median [earnings retention $$\div$$ debt plus equity value at peak] 0.606 0.790 5. Median dividends $$\div$$ dividends plus earnings retention 0.251 0.272 6. Median asset growth during deleveraging 79.9% 190.3% 7. Percentage of firms that increase dividends while deleveraging 100.0% 66.1% 8. Number of firms 1,711 802 9. Percentage of all firms with 5 or more years of post-peak data 38.2% 17.9% Actions taken while deleveraging: Increase dividends Increase debt 1. Median market leverage (ML) at $$\qquad$$ peak 0.582 0.652 $$\qquad$$ trough 0.082 0.169 $$\qquad$$ trough if debt repaid and cash stockpiled with dividend increases 0.008 0.056 2. Median Cash/TA ratio at $$\qquad$$ peak ML 0.042 0.040 $$\qquad$$ trough ML 0.089 0.052 $$\qquad$$ trough if debt repaid and cash stockpiled with dividend increases 0.120 0.073 3. Percentage of firms with negative net debt at $$\qquad$$ trough ML 47.0% 18.5% $$\qquad$$ trough if debt repaid and cash stockpiled with dividend increases 64.3% 46.3% 4. Median [earnings retention $$\div$$ debt plus equity value at peak] 0.606 0.790 5. Median dividends $$\div$$ dividends plus earnings retention 0.251 0.272 6. Median asset growth during deleveraging 79.9% 190.3% 7. Percentage of firms that increase dividends while deleveraging 100.0% 66.1% 8. Number of firms 1,711 802 9. Percentage of all firms with 5 or more years of post-peak data 38.2% 17.9% The samples of dividend-increasing and debt-increasing firms in this table are from the 4,476 firms in the baseline sample with 5 or more years of post-peak data on Compustat. Dividend-increasing firms are those whose average annual dividend payment from the year after peak to the trough year is larger than the dividend payment in the peak year. The last item in rows 1, 2, and 3 reports the hypothetical value of the variable in question that would have obtained at the trough if (1) the firm first used all of the cash that had been allocated to dividend increases to repay debt and (2) if any of the latter cash were left over after paying debt down to zero, the remainder would be placed in cash balances. The variable in row 5 restricts attention to firms with positive retained earnings during the deleveraging episode. The sample in the far-right column of this table is the same as the sample analyzed in the far-right column of Panel A of Table 3. Among dividend-increasing firms, median ML declines from 0.582 at the peak to 0.082 at the trough, but would have been essentially zero (0.008) if cash actually allocated to dividend increases had been used to repay debt (row 1, first column). Although the median Cash/TA ratio increases from 0.042 at the ML peak to 0.089 at the trough, it would have increased to 0.120 if the firm had parked in cash balances the money that was left over after using the cash from dividend increases to pay off all debt (row 2). The incidence of firms with negative net debt at the trough would have been 64.3% instead of the actual 47.0% incidence that obtained after the dividend increases were distributed to shareholders (row 3). These firms typically had high retained earnings while deleveraging and substantial asset growth (rows 4 and 6), with total dividend payments (not just the increases) that constitute about a quarter of the sum of total dividends and retained earnings during the deleveraging episode (row 5). The situation is much the same among firms that increase debt while deleveraging, with dividend increases accounting for the relatively high ML ratio for the median firm at the outcome of deleveraging. For debt-increasing firms, median ML at the trough is 0.169, but it would have been 0.056 if dividend increases had been used to repay debt (row 1, second column), and the median Cash/TA ratio would have been 0.073 rather than 0.052 (row 2). These firms generally expand substantially (rows 4 and 6), which ordinarily would plausibly explain why they take on more debt while deleveraging. However, absent their strong tendency to increase dividend distributions (row 7), the typical ML ratio would have been markedly lower at the post-peak trough, and the typical Cash/TA ratio would have been higher. In sum, many sample firms effectively trade off rebuilding flexibility through ML reductions and cash-balance build ups in order to deliver increasing payouts to shareholders. 4. Heterogeneity in Deleveraging: Basic Findings The stand-out regularity in Sections 2 and 3 is that the typical firm proactively deleverages from peak to near-zero ML. This regularity is consistent with financial flexibility-based theories of capital structure, which indicate that highly levered firms have incentives to deleverage to conservative capital structures to restore the option to borrow. At the same time, however, our data show substantial cross-firm heterogeneity in deleveraging outcomes, with a nontrivial minority of firms not attaining low leverage. The question then becomes whether such attenuated deleveraging reflects other relevant leverage-related factors or whether it is indicative of disinterest in restoring ample financial flexibility. For assessing flexibility-based theories, this section’s key finding is that cases in which firms fail to attain low ML often reflect financial distress, which inherently impedes deleveraging, or decisions to merge into another firm, which of course imply there is no longer a meaningful opportunity for the acquired entity to deleverage. Deleveraging episodes that are attenuated for these reasons are consistent with flexibility-related incentives to deleverage to near-zero ML that are thwarted or dominated by other empirically plausible considerations. 4.1 Some basic facts about deleveraging and financial trouble Table 6 reports Altman z-scores that gauge the extent of financial trouble for the baseline sample, and for subsamples of 4,476 firms with five or more years of post-peak data on Compustat and 5,390 firms with one to four years of post-peak data. z-scores below 1.81 are generally interpreted as indicating a firm in distress that is a likely candidate for bankruptcy. z-scores above 2.99 are commonly viewed as indicating a firm in safe condition, and z-scores between 1.81 and 2.99 are viewed as difficult-to-assess borderline cases in which there is some chance the firm will be facing serious trouble. Table 6 Deleveraging episodes: Financial distress, recession timing, and industry-leverage trends Number of years of post-peak data: 1 or more years 5 or more years 1 to 4 years 1. Median Altman z-score at peak ML 2.01 2.20 1.81 2. Median Altman z-score at post-peak trough 3.50 4.50 2.63 3. Median rate of return on common stock in $$\qquad$$ peak ML year $$-$$38.0% $$-$$34.8% $$-$$41.3% $$\qquad$$ peak ML year and prior year $$-$$46.2% $$-$$45.6% $$-$$46.7% $$\qquad$$ peak ML year and two prior years $$-$$41.5% $$-$$39.7% $$-$$42.9% 4. Percentage of firms with a loss in year of $$\qquad$$ trough before peak ML 43.3% 36.6% 48.9% $$\qquad$$ peak ML 52.2% 45.5% 57.8% $$\qquad$$ trough after peak ML 42.5% 33.0% 50.8% 5. Median return on assets (ROA) in year of $$\qquad$$ trough before peak ML 0.125 0.141 0.111 $$\qquad$$ peak ML 0.074 0.085 0.060 $$\qquad$$ trough after peak ML 0.117 0.141 0.092 6. Percentage of firms with peak ML during recession 42.4% 45.8% 39.5% 7. Median market leverage (ML) at $$\qquad$$ trough before peak ML 0.048 0.061 0.038 $$\qquad$$ at peak ML 0.491 0.543 0.446 $$\qquad$$ at trough after peak ML 0.088 0.026 0.173 8. Median Industry ML at $$\qquad$$ trough before peak ML 0.156 0.162 0.150 $$\qquad$$ at peak ML 0.219 0.241 0.197 $$\qquad$$ at trough after peak ML 0.165 0.159 0.175 9. Number of firms 9,866 4,476 5,390 Number of years of post-peak data: 1 or more years 5 or more years 1 to 4 years 1. Median Altman z-score at peak ML 2.01 2.20 1.81 2. Median Altman z-score at post-peak trough 3.50 4.50 2.63 3. Median rate of return on common stock in $$\qquad$$ peak ML year $$-$$38.0% $$-$$34.8% $$-$$41.3% $$\qquad$$ peak ML year and prior year $$-$$46.2% $$-$$45.6% $$-$$46.7% $$\qquad$$ peak ML year and two prior years $$-$$41.5% $$-$$39.7% $$-$$42.9% 4. Percentage of firms with a loss in year of $$\qquad$$ trough before peak ML 43.3% 36.6% 48.9% $$\qquad$$ peak ML 52.2% 45.5% 57.8% $$\qquad$$ trough after peak ML 42.5% 33.0% 50.8% 5. Median return on assets (ROA) in year of $$\qquad$$ trough before peak ML 0.125 0.141 0.111 $$\qquad$$ peak ML 0.074 0.085 0.060 $$\qquad$$ trough after peak ML 0.117 0.141 0.092 6. Percentage of firms with peak ML during recession 42.4% 45.8% 39.5% 7. Median market leverage (ML) at $$\qquad$$ trough before peak ML 0.048 0.061 0.038 $$\qquad$$ at peak ML 0.491 0.543 0.446 $$\qquad$$ at trough after peak ML 0.088 0.026 0.173 8. Median Industry ML at $$\qquad$$ trough before peak ML 0.156 0.162 0.150 $$\qquad$$ at peak ML 0.219 0.241 0.197 $$\qquad$$ at trough after peak ML 0.165 0.159 0.175 9. Number of firms 9,866 4,476 5,390 Market leverage (ML) is the ratio of the book value of debt to the sum of the book value of debt and the market value of equity. Peak leverage is the maximum value of ML over a firm’s entire time in the sample. The trough before (after) peak is the lowest value of a firm’s ML that comes prior to (subsequent to) its peak. Results for the full baseline sample are in the first column. The next two columns partition the baseline sample into firms with (1) at least five years of post-peak data on Compustat and (2) between one and four years of post-peak data. In row 4, a loss is a negative value of the firm’s earnings before extraordinary items (EBEI). The return on assets (ROA) figure in row 5 is the Rajan and Zingales measure of profitability and equals EBITDA divided by total assets. Since EBEI is generally lower than EBITDA (due, at a minimum, to netting out interest), there is no paradox in the peak year findings that most firms in the baseline sample report losses (negative EBEI) yet ROA is positive for the median firm. Altman z-scores below 1.81 are generally interpreted as indicating a firm in distress that is a likely candidate for bankruptcy. z-scores above 2.99 are generally viewed as indicating the firm is safe, while z-scores between 1.81 and 2.99 are viewed as indicating that it is difficult to tell if the firm will experience distress. A deleveraging episode is classified as recession related if the calendar date of the balance sheet that has peak ML is within six months of the closest month classified as a recession by the NBER. To obtain the industry median ML ratios reported in row 8, we use the approach of Denis and McKeon (2012, p.1902), with the exception that we require a minimum of five other peer firms in the same 4-digit SIC industry (rather than their minimum of 10) in order to use the 4-digit peer group. If that condition is violated, we use a 3-digit industry peer group, provided there are at least five peer firms in the relevant 3-digit industry. If the latter condition is not satisfied, we use a 2-digit peer group. Table 6 Deleveraging episodes: Financial distress, recession timing, and industry-leverage trends Number of years of post-peak data: 1 or more years 5 or more years 1 to 4 years 1. Median Altman z-score at peak ML 2.01 2.20 1.81 2. Median Altman z-score at post-peak trough 3.50 4.50 2.63 3. Median rate of return on common stock in $$\qquad$$ peak ML year $$-$$38.0% $$-$$34.8% $$-$$41.3% $$\qquad$$ peak ML year and prior year $$-$$46.2% $$-$$45.6% $$-$$46.7% $$\qquad$$ peak ML year and two prior years $$-$$41.5% $$-$$39.7% $$-$$42.9% 4. Percentage of firms with a loss in year of $$\qquad$$ trough before peak ML 43.3% 36.6% 48.9% $$\qquad$$ peak ML 52.2% 45.5% 57.8% $$\qquad$$ trough after peak ML 42.5% 33.0% 50.8% 5. Median return on assets (ROA) in year of $$\qquad$$ trough before peak ML 0.125 0.141 0.111 $$\qquad$$ peak ML 0.074 0.085 0.060 $$\qquad$$ trough after peak ML 0.117 0.141 0.092 6. Percentage of firms with peak ML during recession 42.4% 45.8% 39.5% 7. Median market leverage (ML) at $$\qquad$$ trough before peak ML 0.048 0.061 0.038 $$\qquad$$ at peak ML 0.491 0.543 0.446 $$\qquad$$ at trough after peak ML 0.088 0.026 0.173 8. Median Industry ML at $$\qquad$$ trough before peak ML 0.156 0.162 0.150 $$\qquad$$ at peak ML 0.219 0.241 0.197 $$\qquad$$ at trough after peak ML 0.165 0.159 0.175 9. Number of firms 9,866 4,476 5,390 Number of years of post-peak data: 1 or more years 5 or more years 1 to 4 years 1. Median Altman z-score at peak ML 2.01 2.20 1.81 2. Median Altman z-score at post-peak trough 3.50 4.50 2.63 3. Median rate of return on common stock in $$\qquad$$ peak ML year $$-$$38.0% $$-$$34.8% $$-$$41.3% $$\qquad$$ peak ML year and prior year $$-$$46.2% $$-$$45.6% $$-$$46.7% $$\qquad$$ peak ML year and two prior years $$-$$41.5% $$-$$39.7% $$-$$42.9% 4. Percentage of firms with a loss in year of $$\qquad$$ trough before peak ML 43.3% 36.6% 48.9% $$\qquad$$ peak ML 52.2% 45.5% 57.8% $$\qquad$$ trough after peak ML 42.5% 33.0% 50.8% 5. Median return on assets (ROA) in year of $$\qquad$$ trough before peak ML 0.125 0.141 0.111 $$\qquad$$ peak ML 0.074 0.085 0.060 $$\qquad$$ trough after peak ML 0.117 0.141 0.092 6. Percentage of firms with peak ML during recession 42.4% 45.8% 39.5% 7. Median market leverage (ML) at $$\qquad$$ trough before peak ML 0.048 0.061 0.038 $$\qquad$$ at peak ML 0.491 0.543 0.446 $$\qquad$$ at trough after peak ML 0.088 0.026 0.173 8. Median Industry ML at $$\qquad$$ trough before peak ML 0.156 0.162 0.150 $$\qquad$$ at peak ML 0.219 0.241 0.197 $$\qquad$$ at trough after peak ML 0.165 0.159 0.175 9. Number of firms 9,866 4,476 5,390 Market leverage (ML) is the ratio of the book value of debt to the sum of the book value of debt and the market value of equity. Peak leverage is the maximum value of ML over a firm’s entire time in the sample. The trough before (after) peak is the lowest value of a firm’s ML that comes prior to (subsequent to) its peak. Results for the full baseline sample are in the first column. The next two columns partition the baseline sample into firms with (1) at least five years of post-peak data on Compustat and (2) between one and four years of post-peak data. In row 4, a loss is a negative value of the firm’s earnings before extraordinary items (EBEI). The return on assets (ROA) figure in row 5 is the Rajan and Zingales measure of profitability and equals EBITDA divided by total assets. Since EBEI is generally lower than EBITDA (due, at a minimum, to netting out interest), there is no paradox in the peak year findings that most firms in the baseline sample report losses (negative EBEI) yet ROA is positive for the median firm. Altman z-scores below 1.81 are generally interpreted as indicating a firm in distress that is a likely candidate for bankruptcy. z-scores above 2.99 are generally viewed as indicating the firm is safe, while z-scores between 1.81 and 2.99 are viewed as indicating that it is difficult to tell if the firm will experience distress. A deleveraging episode is classified as recession related if the calendar date of the balance sheet that has peak ML is within six months of the closest month classified as a recession by the NBER. To obtain the industry median ML ratios reported in row 8, we use the approach of Denis and McKeon (2012, p.1902), with the exception that we require a minimum of five other peer firms in the same 4-digit SIC industry (rather than their minimum of 10) in order to use the 4-digit peer group. If that condition is violated, we use a 3-digit industry peer group, provided there are at least five peer firms in the relevant 3-digit industry. If the latter condition is not satisfied, we use a 2-digit peer group. The Altman z-scores in Table 6 indicate that most firms exhibit nontrivial signs of distress at peak ML (row 1). It thus makes sense that most managers would make decisions that reduce ML, as Section 3 indicates they do. The net result is that, after deleveraging, z-scores indicate that most firms are in safe condition (row 2). Consistent with distress around ML peaks, most firms have large negative stock returns in the year they attain peak ML, with equity values falling a modest amount the prior year, and not declining the year before that (row 3). Most firms report negative earnings in the year of peak ML, with a somewhat stronger tendency toward peak-year losses among firms with one-to-four years of post-peak data (row 4). Most also have lower ROA in the peak ML year than they had at the prior trough, with ROA recovering to a large degree by the trough after the peak (row 5). The trough-peak-trough reversal pattern of losses and ROA indicates transitory problems for most firms around peak ML. We also find that 42.4% of ML peaks occur during NBER recessions (row 6, Table 6), which is consistent with the view that the financial difficulties at the peak are often both transitory and not fully idiosyncratic to individual firms. Figure 5 plots the calendar-time incidence of ML peaks for all firms in the baseline sample and for firms with at least 20 years of data, with gray background identifying recession periods. The 1974 recession stands out strongly in the figure, accounting for almost 8.0% of the ML peaks in the baseline sample and almost 15.0% of the peaks among firms with 20 or more years of data. The 15.0% incidence is remarkably high and noteworthy because this subsample of firms contains many of the most prominent nonfinancial companies. Figure 5 View largeDownload slide Calendar timing of market-leverage peaks Market leverage (ML) equals the book value of debt divided by the sum of the book value of debt and the market value of equity. The figure reports the percentage of firms for which the peak ML came in the specified calendar year. Recession timing is marked with gray vertical shading of the background in the figure and is based on the NBER’s classification of recessions. Since our data are grouped by year on Compustat, while the NBER classifies recessions based on months, the figure portrays a given year as a recession period if at least one month in that year is classified as such by the NBER. (As a result, the figure shows no separation between the 1980 and the 1981 recessions, despite the brief respite noted in the NBER’s monthly data.) See Table 6 for details on how we classify each deleveraging episode as related (or not) to recessions. The baseline sample has 14,196 firms with at least two years of data on Compustat. The 20-year plus sample contains the subset of 2,738 firms with at least 20 years of data on Compustat. These samples exclude cases in which ML takes the same value in all years on Compustat (since such firms do not have an economically meaningful peak ML). For example, the 20-year plus sample excludes eight firms that have ML $$=$$ 0.000 in all years. Figure 5 View largeDownload slide Calendar timing of market-leverage peaks Market leverage (ML) equals the book value of debt divided by the sum of the book value of debt and the market value of equity. The figure reports the percentage of firms for which the peak ML came in the specified calendar year. Recession timing is marked with gray vertical shading of the background in the figure and is based on the NBER’s classification of recessions. Since our data are grouped by year on Compustat, while the NBER classifies recessions based on months, the figure portrays a given year as a recession period if at least one month in that year is classified as such by the NBER. (As a result, the figure shows no separation between the 1980 and the 1981 recessions, despite the brief respite noted in the NBER’s monthly data.) See Table 6 for details on how we classify each deleveraging episode as related (or not) to recessions. The baseline sample has 14,196 firms with at least two years of data on Compustat. The 20-year plus sample contains the subset of 2,738 firms with at least 20 years of data on Compustat. These samples exclude cases in which ML takes the same value in all years on Compustat (since such firms do not have an economically meaningful peak ML). For example, the 20-year plus sample excludes eight firms that have ML $$=$$ 0.000 in all years. While the recession findings point to material comovement across firms in deleveraging, such comovement is typically modest at the industry level. Peer firms’ ML ratios do tend to increase when sample firms are increasing ML toward the peak. They also tend to decrease when sample firms subsequently deleverage. However, the changes in industry ML ratios are small relative to the changes in sample firms’ ML ratios (compare rows 7 and 8 in Table 6). 4.2 Heterogeneity in the extent of financial trouble among deleveraging firms Table 7 documents the cross-firm distribution of ML ratios at the peak (panel A) and at the post-peak trough (panel B), with row 4 reporting the median Altman z-score for each ML category in the columns. The table reports data for the baseline sample partitioned into firms with five or more years of post-peak data on Compustat and those with between one and four years of data. Table 7 Distributions of market leverage ML ratios at the peak and at the post-peak trough: Baseline sample partitioned by number of years of post-peak data provided in the Compustat file A. ML at the peak Peak ML in specified range: All firms 0.000 to 0.100 0.100 to 0.200 0.200 to 0.300 0.300 to 0.400 0.400 to 0.500 0.500 to 0.600 0.600 to 0.700 0.700 to 0.800 0.800 to 0.900 0.900 to 0.999 1. Number of firms $$\qquad$$ 5$$+$$ years of post-peak data 4,476 901 562 472 520 542 479 524 487 473 430 $$\qquad$$ 1-to-4 years of post-peak data 5,390 512 320 338 419 433 505 586 545 477 341 2. Percentage of firms $$\qquad$$ 5$$+$$ years of post-peak data 100.0% 11.4% 7.1% 7.6% 9.3% 9.7% 11.3% 13.1% 12.2% 10.7% 7.6% $$\qquad$$ 1-to-4 years of post-peak data 100.0% 16.7% 10.4% 8.8% 9.6% 10.1% 8.9% 9.7% 9.0% 8.8% 8.0% 3. Cumulative percentage of firms $$\qquad$$ 5$$+$$ years of post-peak data – 11.4% 18.5% 26.1% 35.4% 45.1% 56.4% 69.5% 81.7% 92.4% 100.0% $$\qquad$$ 1-to-4 years of post-peak data – 16.7% 27.1% 35.9% 45.5% 55.6% 64.5% 74.2% 83.2% 92.0% 100.0% 4. Median Altman z-score at peak $$\qquad$$ 5$$+$$ years of post-peak data 2.20 5.85 3.48 2.63 2.61 2.19 2.24 2.04 2.02 1.46 0.63 $$\qquad$$ 1-to-4 years of post-peak data 1.81 4.59 2.77 2.27 2.16 1.98 1.60 1.51 1.41 1.12 0.16 B. ML at the trough after peak Post-peak ML in specified range: 0.000 0.000 to 0.100 0.100 to 0.200 0.200 to 0.300 0.300 to 0.400 0.400 to 0.500 0.500 to 0.600 0.600 to 0.700 0.700 to 0.800 0.800 to 0.900 0.900 to 0.999 1. Number of firms $$\qquad$$ 5$$+$$ years of post-peak data 1,488 1,447 635 380 225 133 93 41 22 9 3 $$\qquad$$ 1-to-4 years of post-peak data 764 1,422 729 592 503 360 323 259 204 137 97 2. Percentage of firms $$\qquad$$ 5$$+$$ years of post-peak data 33.2% 32.3% 14.2% 8.5% 5.0% 3.0% 2.1% 0.9% 0.5% 0.2% 0.1% $$\qquad$$ 1-to-4 years of post-peak data 14.2% 26.4% 13.5% 11.0% 9.3% 6.7% 6.0% 4.8% 3.8% 2.5% 1.8% 3. Cumulative percentage of firms $$\qquad$$ 5$$+$$ years of post-peak data 33.2% 65.5% 79.7% 88.2% 93.2% 96.2% 98.3% 99.2% 99.7% 99.9% 100.0% $$\qquad$$ 1-to-4 years of post-peak data 14.2% 40.6% 54.1% 65.1% 74.4% 81.1% 87.1% 91.9% 95.7% 98.2% 100.0% 4. Median Altman z-score at post-peak trough $$\qquad$$ 5$$+$$ years of post-peak data 5.75 5.81 3.89 3.36 3.04 2.48 2.26 1.39 2.06 1.77 $$-$$5.07 $$\qquad$$ 1-to-4 years of post-peak data 3.88 4.67 3.17 2.67 2.37 1.87 1.80 1.49 1.27 0.65 $$-$$0.68 A. ML at the peak Peak ML in specified range: All firms 0.000 to 0.100 0.100 to 0.200 0.200 to 0.300 0.300 to 0.400 0.400 to 0.500 0.500 to 0.600 0.600 to 0.700 0.700 to 0.800 0.800 to 0.900 0.900 to 0.999 1. Number of firms $$\qquad$$ 5$$+$$ years of post-peak data 4,476 901 562 472 520 542 479 524 487 473 430 $$\qquad$$ 1-to-4 years of post-peak data 5,390 512 320 338 419 433 505 586 545 477 341 2. Percentage of firms $$\qquad$$ 5$$+$$ years of post-peak data 100.0% 11.4% 7.1% 7.6% 9.3% 9.7% 11.3% 13.1% 12.2% 10.7% 7.6% $$\qquad$$ 1-to-4 years of post-peak data 100.0% 16.7% 10.4% 8.8% 9.6% 10.1% 8.9% 9.7% 9.0% 8.8% 8.0% 3. Cumulative percentage of firms $$\qquad$$ 5$$+$$ years of post-peak data – 11.4% 18.5% 26.1% 35.4% 45.1% 56.4% 69.5% 81.7% 92.4% 100.0% $$\qquad$$ 1-to-4 years of post-peak data – 16.7% 27.1% 35.9% 45.5% 55.6% 64.5% 74.2% 83.2% 92.0% 100.0% 4. Median Altman z-score at peak $$\qquad$$ 5$$+$$ years of post-peak data 2.20 5.85 3.48 2.63 2.61 2.19 2.24 2.04 2.02 1.46 0.63 $$\qquad$$ 1-to-4 years of post-peak data 1.81 4.59 2.77 2.27 2.16 1.98 1.60 1.51 1.41 1.12 0.16 B. ML at the trough after peak Post-peak ML in specified range: 0.000 0.000 to 0.100 0.100 to 0.200 0.200 to 0.300 0.300 to 0.400 0.400 to 0.500 0.500 to 0.600 0.600 to 0.700 0.700 to 0.800 0.800 to 0.900 0.900 to 0.999 1. Number of firms $$\qquad$$ 5$$+$$ years of post-peak data 1,488 1,447 635 380 225 133 93 41 22 9 3 $$\qquad$$ 1-to-4 years of post-peak data 764 1,422 729 592 503 360 323 259 204 137 97 2. Percentage of firms $$\qquad$$ 5$$+$$ years of post-peak data 33.2% 32.3% 14.2% 8.5% 5.0% 3.0% 2.1% 0.9% 0.5% 0.2% 0.1% $$\qquad$$ 1-to-4 years of post-peak data 14.2% 26.4% 13.5% 11.0% 9.3% 6.7% 6.0% 4.8% 3.8% 2.5% 1.8% 3. Cumulative percentage of firms $$\qquad$$ 5$$+$$ years of post-peak data 33.2% 65.5% 79.7% 88.2% 93.2% 96.2% 98.3% 99.2% 99.7% 99.9% 100.0% $$\qquad$$ 1-to-4 years of post-peak data 14.2% 40.6% 54.1% 65.1% 74.4% 81.1% 87.1% 91.9% 95.7% 98.2% 100.0% 4. Median Altman z-score at post-peak trough $$\qquad$$ 5$$+$$ years of post-peak data 5.75 5.81 3.89 3.36 3.04 2.48 2.26 1.39 2.06 1.77 $$-$$5.07 $$\qquad$$ 1-to-4 years of post-peak data 3.88 4.67 3.17 2.67 2.37 1.87 1.80 1.49 1.27 0.65 $$-$$0.68 Market leverage (ML) is book debt divided by the sum of book debt and the market value of equity. Peak leverage is the maximum ML over a firm’s time in the sample. ML at the post-peak trough is the lowest value of ML that comes after the peak. The table partitions the baseline sample of 9,866 firms into the 4,476 firms with five or more years of post-peak data on Compustat and the 5,390 firms with between one and four years of post-peak data. Altman z-scores below 1.81 are generally interpreted as indicating a firm in distress that is a likely candidate for bankruptcy. z-scores above 2.99 are generally viewed as indicating a firm that is safe, and z-scores between 1.81 and 2.99 are viewed as indicating that it is difficult to tell if the firm will experience distress. Table 7 Distributions of market leverage ML ratios at the peak and at the post-peak trough: Baseline sample partitioned by number of years of post-peak data provided in the Compustat file A. ML at the peak Peak ML in specified range: All firms 0.000 to 0.100 0.100 to 0.200 0.200 to 0.300 0.300 to 0.400 0.400 to 0.500 0.500 to 0.600 0.600 to 0.700 0.700 to 0.800 0.800 to 0.900 0.900 to 0.999 1. Number of firms $$\qquad$$ 5$$+$$ years of post-peak data 4,476 901 562 472 520 542 479 524 487 473 430 $$\qquad$$ 1-to-4 years of post-peak data 5,390 512 320 338 419 433 505 586 545 477 341 2. Percentage of firms $$\qquad$$ 5$$+$$ years of post-peak data 100.0% 11.4% 7.1% 7.6% 9.3% 9.7% 11.3% 13.1% 12.2% 10.7% 7.6% $$\qquad$$ 1-to-4 years of post-peak data 100.0% 16.7% 10.4% 8.8% 9.6% 10.1% 8.9% 9.7% 9.0% 8.8% 8.0% 3. Cumulative percentage of firms $$\qquad$$ 5$$+$$ years of post-peak data – 11.4% 18.5% 26.1% 35.4% 45.1% 56.4% 69.5% 81.7% 92.4% 100.0% $$\qquad$$ 1-to-4 years of post-peak data – 16.7% 27.1% 35.9% 45.5% 55.6% 64.5% 74.2% 83.2% 92.0% 100.0% 4. Median Altman z-score at peak $$\qquad$$ 5$$+$$ years of post-peak data 2.20 5.85 3.48 2.63 2.61 2.19 2.24 2.04 2.02 1.46 0.63 $$\qquad$$ 1-to-4 years of post-peak data 1.81 4.59 2.77 2.27 2.16 1.98 1.60 1.51 1.41 1.12 0.16 B. ML at the trough after peak Post-peak ML in specified range: 0.000 0.000 to 0.100 0.100 to 0.200 0.200 to 0.300 0.300 to 0.400 0.400 to 0.500 0.500 to 0.600 0.600 to 0.700 0.700 to 0.800 0.800 to 0.900 0.900 to 0.999 1. Number of firms $$\qquad$$ 5$$+$$ years of post-peak data 1,488 1,447 635 380 225 133 93 41 22 9 3 $$\qquad$$ 1-to-4 years of post-peak data 764 1,422 729 592 503 360 323 259 204 137 97 2. Percentage of firms $$\qquad$$ 5$$+$$ years of post-peak data 33.2% 32.3% 14.2% 8.5% 5.0% 3.0% 2.1% 0.9% 0.5% 0.2% 0.1% $$\qquad$$ 1-to-4 years of post-peak data 14.2% 26.4% 13.5% 11.0% 9.3% 6.7% 6.0% 4.8% 3.8% 2.5% 1.8% 3. Cumulative percentage of firms $$\qquad$$ 5$$+$$ years of post-peak data 33.2% 65.5% 79.7% 88.2% 93.2% 96.2% 98.3% 99.2% 99.7% 99.9% 100.0% $$\qquad$$ 1-to-4 years of post-peak data 14.2% 40.6% 54.1% 65.1% 74.4% 81.1% 87.1% 91.9% 95.7% 98.2% 100.0% 4. Median Altman z-score at post-peak trough $$\qquad$$ 5$$+$$ years of post-peak data 5.75 5.81 3.89 3.36 3.04 2.48 2.26 1.39 2.06 1.77 $$-$$5.07 $$\qquad$$ 1-to-4 years of post-peak data 3.88 4.67 3.17 2.67 2.37 1.87 1.80 1.49 1.27 0.65 $$-$$0.68 A. ML at the peak Peak ML in specified range: All firms 0.000 to 0.100 0.100 to 0.200 0.200 to 0.300 0.300 to 0.400 0.400 to 0.500 0.500 to 0.600 0.600 to 0.700 0.700 to 0.800 0.800 to 0.900 0.900 to 0.999 1. Number of firms $$\qquad$$ 5$$+$$ years of post-peak data 4,476 901 562 472 520 542 479 524 487 473 430 $$\qquad$$ 1-to-4 years of post-peak data 5,390 512 320 338 419 433 505 586 545 477 341 2. Percentage of firms $$\qquad$$ 5$$+$$ years of post-peak data 100.0% 11.4% 7.1% 7.6% 9.3% 9.7% 11.3% 13.1% 12.2% 10.7% 7.6% $$\qquad$$ 1-to-4 years of post-peak data 100.0% 16.7% 10.4% 8.8% 9.6% 10.1% 8.9% 9.7% 9.0% 8.8% 8.0% 3. Cumulative percentage of firms $$\qquad$$ 5$$+$$ years of post-peak data – 11.4% 18.5% 26.1% 35.4% 45.1% 56.4% 69.5% 81.7% 92.4% 100.0% $$\qquad$$ 1-to-4 years of post-peak data – 16.7% 27.1% 35.9% 45.5% 55.6% 64.5% 74.2% 83.2% 92.0% 100.0% 4. Median Altman z-score at peak $$\qquad$$ 5$$+$$ years of post-peak data 2.20 5.85 3.48 2.63 2.61 2.19 2.24 2.04 2.02 1.46 0.63 $$\qquad$$ 1-to-4 years of post-peak data 1.81 4.59 2.77 2.27 2.16 1.98 1.60 1.51 1.41 1.12 0.16 B. ML at the trough after peak Post-peak ML in specified range: 0.000 0.000 to 0.100 0.100 to 0.200 0.200 to 0.300 0.300 to 0.400 0.400 to 0.500 0.500 to 0.600 0.600 to 0.700 0.700 to 0.800 0.800 to 0.900 0.900 to 0.999 1. Number of firms $$\qquad$$ 5$$+$$ years of post-peak data 1,488 1,447 635 380 225 133 93 41 22 9 3 $$\qquad$$ 1-to-4 years of post-peak data 764 1,422 729 592 503 360 323 259 204 137 97 2. Percentage of firms $$\qquad$$ 5$$+$$ years of post-peak data 33.2% 32.3% 14.2% 8.5% 5.0% 3.0% 2.1% 0.9% 0.5% 0.2% 0.1% $$\qquad$$ 1-to-4 years of post-peak data 14.2% 26.4% 13.5% 11.0% 9.3% 6.7% 6.0% 4.8% 3.8% 2.5% 1.8% 3. Cumulative percentage of firms $$\qquad$$ 5$$+$$ years of post-peak data 33.2% 65.5% 79.7% 88.2% 93.2% 96.2% 98.3% 99.2% 99.7% 99.9% 100.0% $$\qquad$$ 1-to-4 years of post-peak data 14.2% 40.6% 54.1% 65.1% 74.4% 81.1% 87.1% 91.9% 95.7% 98.2% 100.0% 4. Median Altman z-score at post-peak trough $$\qquad$$ 5$$+$$ years of post-peak data 5.75 5.81 3.89 3.36 3.04 2.48 2.26 1.39 2.06 1.77 $$-$$5.07 $$\qquad$$ 1-to-4 years of post-peak data 3.88 4.67 3.17 2.67 2.37 1.87 1.80 1.49 1.27 0.65 $$-$$0.68 Market leverage (ML) is book debt divided by the sum of book debt and the market value of equity. Peak leverage is the maximum ML over a firm’s time in the sample. ML at the post-peak trough is the lowest value of ML that comes after the peak. The table partitions the baseline sample of 9,866 firms into the 4,476 firms with five or more years of post-peak data on Compustat and the 5,390 firms with between one and four years of post-peak data. Altman z-scores below 1.81 are generally interpreted as indicating a firm in distress that is a likely candidate for bankruptcy. z-scores above 2.99 are generally viewed as indicating a firm that is safe, and z-scores between 1.81 and 2.99 are viewed as indicating that it is difficult to tell if the firm will experience distress. Panel A of Table 7 indicates that the one-to-four-years group tends to have a higher percentage of firms with low peak ML ratios (rows 2 and 3). However, holding the level of peak ML fixed, this group tends to show stronger signs of financial trouble than the five-plus-years group (row 4). Relatively few firms continue to have very high ML ratios at the post-peak trough, with those that remain highly levered mainly concentrated among firms with four or fewer years of post-peak data (row 1, panel B, Table 7). In the five-plus-years group, only 4.8% of firms fail to deleverage to ML below 0.500, while 18.9% of the one-to-four-years group fail to do so (row 3). Median Altman z-scores for the latter firms are uniformly indicative of serious financial trouble (row 4). The overall picture from Table 7 is that, at the end of their deleveraging episodes, a nontrivial subset of firms has serious financial troubles and is still highly levered, with such cases concentrated among firms with four or fewer years of post-peak data. 4.3 Attenuated deleveraging and delisting due to financial distress and mergers The idea that financially troubled firms find it difficult to restructure their debt outside of bankruptcy court is supported by numerous studies, including Gilson, John, and Lang (1990), Asquith, Gertner, and Scharfstein (1994), and Gilson (1997). For example, in Gilson’s (1997) sample of out-of-court debt restructurings, the median firm’s long-term debt ratio (long-term debt/(long-term debt $$+$$ equity value)) declines by only 0.060—from 0.700 to 0.640—with deleveraging thwarted to the point that 35% of these firms later undergo further debt restructuring. Table 8 documents distress-related attenuated deleveraging—to an ML ratio that is lower than in Gilson’s study, but still well above zero—for many firms in our baseline sample. The table also reports nontrivial attenuated deleveraging by many firms that are acquired in their peak ML year, or soon after. The table indicates that 10.4% of firms in the baseline sample were delisted (per CRSP) because of financial distress in the four years after attaining peak ML, while another 11.5% were delisted due to distress almost immediately after attaining peak ML (row 2). The analogous sample incidences for firms delisted due to acquisition are 14.0% and 7.6% (row 2). We focus on the first and third columns of the table, which examine the distressed and merger delists that occur in the four years after peak ML. We focus on these firms because Compustat has data to gauge the extent of their post-peak deleveraging. Table 8 Attenuated deleveraging: Firms delisted because of financial distress or merger Financial distress delists Merger delists 1 to 4 years in sample after peak ML Peak ML occurs in last year in sample 1 to 4 years in sample after peak ML Peak ML occurs in last year in sample 1. Number of firms 1,482 1,636 1,976 1,077 2. Percentage relative to 14,196 firms in baseline sample 10.4% 11.5% 14.0% 7.6% 3. Median years from peak market leverage (ML) to trough 1 na 1 na 4. Median ML at trough before peak 0.042 0.034 0.046 0.017 5. Median ML at peak 0.523 0.627 0.382 0.347 6. Median ML at trough after peak 0.181 na 0.182 na 7. Median Cash/TA at trough before ML peak 0.161 0.152 0.129 0.158 8. Median Cash/TA at peak ML 0.055 0.039 0.059 0.062 9. Median Cash/TA at trough after peak 0.068 na 0.085 na 10. Median Altman z-score at ML peak 0.49 $$-$$0.95 2.58 2.07 11. Median Altman z-score at post-peak trough 0.49 na 3.35 na 12. Percentage of firms with a loss in peak ML year 82.1% 92.6 39.7 52.2 13. Median percentage change in debt in peak ML year 12.5% 12.5% 23.8% 41.3% 14. Median total stock market return in peak ML year $$-$$54.2% $$-$$63.1% $$-$$30.3% $$-$$30.4% 15. Median total stock market return in peak ML and prior year $$-$$67.1% $$-$$78.6% $$-$$31.0% $$-$$38.3% 16. Median changes over period from peak ML to later trough: $$\qquad$$ Percentage change in debt $$-$$52.8% na $$-$$24.2% na $$\qquad$$ Earnings retained $$\div$$ debt plus equity value at peak $$-$$0.165 na 0.040 na $$\qquad$$ Percentage change in shares outstanding 2.9% na 1.2% na $$\qquad$$ Percentage change in total assets $$-$$21.5% na 5.2% na Financial distress delists Merger delists 1 to 4 years in sample after peak ML Peak ML occurs in last year in sample 1 to 4 years in sample after peak ML Peak ML occurs in last year in sample 1. Number of firms 1,482 1,636 1,976 1,077 2. Percentage relative to 14,196 firms in baseline sample 10.4% 11.5% 14.0% 7.6% 3. Median years from peak market leverage (ML) to trough 1 na 1 na 4. Median ML at trough before peak 0.042 0.034 0.046 0.017 5. Median ML at peak 0.523 0.627 0.382 0.347 6. Median ML at trough after peak 0.181 na 0.182 na 7. Median Cash/TA at trough before ML peak 0.161 0.152 0.129 0.158 8. Median Cash/TA at peak ML 0.055 0.039 0.059 0.062 9. Median Cash/TA at trough after peak 0.068 na 0.085 na 10. Median Altman z-score at ML peak 0.49 $$-$$0.95 2.58 2.07 11. Median Altman z-score at post-peak trough 0.49 na 3.35 na 12. Percentage of firms with a loss in peak ML year 82.1% 92.6 39.7 52.2 13. Median percentage change in debt in peak ML year 12.5% 12.5% 23.8% 41.3% 14. Median total stock market return in peak ML year $$-$$54.2% $$-$$63.1% $$-$$30.3% $$-$$30.4% 15. Median total stock market return in peak ML and prior year $$-$$67.1% $$-$$78.6% $$-$$31.0% $$-$$38.3% 16. Median changes over period from peak ML to later trough: $$\qquad$$ Percentage change in debt $$-$$52.8% na $$-$$24.2% na $$\qquad$$ Earnings retained $$\div$$ debt plus equity value at peak $$-$$0.165 na 0.040 na $$\qquad$$ Percentage change in shares outstanding 2.9% na 1.2% na $$\qquad$$ Percentage change in total assets $$-$$21.5% na 5.2% na Market leverage (ML) is book debt divided by the sum of book debt and the market value of equity. Peak leverage is the maximum ML over a firm’s time in the sample. The first column contains the firms with one to four post-peak years of data that were delisted due to financial distress during that period (CRSP delist code in the 400s or 500s). The second column contains the firms with peak ML in their last year on Compustat that were delisted (per CRSP) due to financial distress in the year of or year after peak ML. The third column contains the firms with one to four post-peak years of data that were delisted due to mergers or acquisitions (CRSP delist code in the 200s or 300s). The fourth column contains the firms with peak ML in their last year on Compustat that were delisted due to mergers or acquisitions in the year of or year after peak ML. Altman z-scores below 1.81 are generally interpreted as indicating a firm in distress that is a likely candidate for bankruptcy. z-scores above 2.99 are generally viewed as indicating a firm that is safe, and z-scores between 1.81 and 2.99 are viewed as indicating that it is difficult to tell if the firm will experience distress. na, data are not available. Table 8 Attenuated deleveraging: Firms delisted because of financial distress or merger Financial distress delists Merger delists 1 to 4 years in sample after peak ML Peak ML occurs in last year in sample 1 to 4 years in sample after peak ML Peak ML occurs in last year in sample 1. Number of firms 1,482 1,636 1,976 1,077 2. Percentage relative to 14,196 firms in baseline sample 10.4% 11.5% 14.0% 7.6% 3. Median years from peak market leverage (ML) to trough 1 na 1 na 4. Median ML at trough before peak 0.042 0.034 0.046 0.017 5. Median ML at peak 0.523 0.627 0.382 0.347 6. Median ML at trough after peak 0.181 na 0.182 na 7. Median Cash/TA at trough before ML peak 0.161 0.152 0.129 0.158 8. Median Cash/TA at peak ML 0.055 0.039 0.059 0.062 9. Median Cash/TA at trough after peak 0.068 na 0.085 na 10. Median Altman z-score at ML peak 0.49 $$-$$0.95 2.58 2.07 11. Median Altman z-score at post-peak trough 0.49 na 3.35 na 12. Percentage of firms with a loss in peak ML year 82.1% 92.6 39.7 52.2 13. Median percentage change in debt in peak ML year 12.5% 12.5% 23.8% 41.3% 14. Median total stock market return in peak ML year $$-$$54.2% $$-$$63.1% $$-$$30.3% $$-$$30.4% 15. Median total stock market return in peak ML and prior year $$-$$67.1% $$-$$78.6% $$-$$31.0% $$-$$38.3% 16. Median changes over period from peak ML to later trough: $$\qquad$$ Percentage change in debt $$-$$52.8% na $$-$$24.2% na $$\qquad$$ Earnings retained $$\div$$ debt plus equity value at peak $$-$$0.165 na 0.040 na $$\qquad$$ Percentage change in shares outstanding 2.9% na 1.2% na $$\qquad$$ Percentage change in total assets $$-$$21.5% na 5.2% na Financial distress delists Merger delists 1 to 4 years in sample after peak ML Peak ML occurs in last year in sample 1 to 4 years in sample after peak ML Peak ML occurs in last year in sample 1. Number of firms 1,482 1,636 1,976 1,077 2. Percentage relative to 14,196 firms in baseline sample 10.4% 11.5% 14.0% 7.6% 3. Median years from peak market leverage (ML) to trough 1 na 1 na 4. Median ML at trough before peak 0.042 0.034 0.046 0.017 5. Median ML at peak 0.523 0.627 0.382 0.347 6. Median ML at trough after peak 0.181 na 0.182 na 7. Median Cash/TA at trough before ML peak 0.161 0.152 0.129 0.158 8. Median Cash/TA at peak ML 0.055 0.039 0.059 0.062 9. Median Cash/TA at trough after peak 0.068 na 0.085 na 10. Median Altman z-score at ML peak 0.49 $$-$$0.95 2.58 2.07 11. Median Altman z-score at post-peak trough 0.49 na 3.35 na 12. Percentage of firms with a loss in peak ML year 82.1% 92.6 39.7 52.2 13. Median percentage change in debt in peak ML year 12.5% 12.5% 23.8% 41.3% 14. Median total stock market return in peak ML year $$-$$54.2% $$-$$63.1% $$-$$30.3% $$-$$30.4% 15. Median total stock market return in peak ML and prior year $$-$$67.1% $$-$$78.6% $$-$$31.0% $$-$$38.3% 16. Median changes over period from peak ML to later trough: $$\qquad$$ Percentage change in debt $$-$$52.8% na $$-$$24.2% na $$\qquad$$ Earnings retained $$\div$$ debt plus equity value at peak $$-$$0.165 na 0.040 na $$\qquad$$ Percentage change in shares outstanding 2.9% na 1.2% na $$\qquad$$ Percentage change in total assets $$-$$21.5% na 5.2% na Market leverage (ML) is book debt divided by the sum of book debt and the market value of equity. Peak leverage is the maximum ML over a firm’s time in the sample. The first column contains the firms with one to four post-peak years of data that were delisted due to financial distress during that period (CRSP delist code in the 400s or 500s). The second column contains the firms with peak ML in their last year on Compustat that were delisted (per CRSP) due to financial distress in the year of or year after peak ML. The third column contains the firms with one to four post-peak years of data that were delisted due to mergers or acquisitions (CRSP delist code in the 200s or 300s). The fourth column contains the firms with peak ML in their last year on Compustat that were delisted due to mergers or acquisitions in the year of or year after peak ML. Altman z-scores below 1.81 are generally interpreted as indicating a firm in distress that is a likely candidate for bankruptcy. z-scores above 2.99 are generally viewed as indicating a firm that is safe, and z-scores between 1.81 and 2.99 are viewed as indicating that it is difficult to tell if the firm will experience distress. na, data are not available. Table 8 reports that, among firms that are delisted due to distress, median ML at the peak is 0.523 (row 5), which is close to the analogous 0.543 figure reported in Table 2 for firms with five or more years of post-peak data. Distress delists have a median ML of 0.181 at the post-peak trough (row 6), which is well above the analogous 0.026 figure in Table 2 for firms with five or more years of post-peak data. It is also well above the 0.042 ML ratio that the median firm in this distressed-firm subsample had at the trough before peak ML (row 4). Altman z-scores indicate serious financial trouble at peak ML and at the post-peak trough for distress delists (rows 10 and 11). Consistent with serious trouble, most of these firms lost money in the peak ML year (row 12), and they typically took on little additional debt (row 13) and instead reached peak ML due to a large fall in equity value (rows 14 and 15). The median firm has negative retained earnings while deleveraging and a resultant large contraction in total assets (row 16). Firms delisted due to merger also show attenuated deleveraging, with median ML declining from 0.382 at the peak to 0.182 (rows 5 and 6) rather than to a near-zero ML ratio. However, these firms are much less troubled than the distress delists according to Altman z-scores and other financial distress indicators (rows 10, 11, 12, 14, and 15). After attainment of peak ML, distressed delists and merger delists both tend to have relatively small Cash/TA increases that mirror their modest ML decreases (rows 8 and 9). Firms in both subsamples typically repay a substantial portion of the debt they had outstanding at peak ML (row 16). However, for the median firm among the distress delists, retained earnings actually erode after peak ML and only a small number of new shares are issued, while retention and share issuance are positive, but economically inconsequential for the median merger delist (row 16). For distress delists, the attenuated deleveraging from peak ML is plausibly something that managers would have liked to avoid, but could not because financial troubles impeded further deleveraging. In any case, since delisting (due to distress or merger) eliminates the ability to observe the leverage policies that managers would have pursued absent delisting, it seems unwarranted to view their deleveraging outcomes as representative of the outcomes of firms that remain listed. 4.4 The path to peak ML: Proactive debt issuance and equity-value declines A potentially important source of path dependency in deleveraging concerns how firms arrive at peak ML which, of course, is the start of the deleveraging episodes we study. One might reasonably expect that the leverage paths of firms that proactively choose to move to peak ML differ materially from those of firms that did not choose to move to peak. To assess this issue, we apply the conditions used by Denis and McKeon (2012) to identify firms that reach their peaks by proactively increasing ML by large amounts.6 The underlying premise is that few managers voluntarily pursue declines in their firms’ stock market values and so, in identifying proactive ML increases, one should isolate cases in which debt issuances are large relative to any declines in equity value. Our key findings are in Table 9 which, for brevity, restricts attention to firms with at least five years of post-peak data. Table 9 Deleveraging episodes: Sample partitioned by whether firms proactively move to peak market leverage Movement to peak ML Median value of Proactive Other 1. Market leverage (ML) at peak 0.489 0.606 2. ML at post-peak trough 0.028 0.037 3. Percentage change in debt in peak ML year 93.4% $$-$$0.5% 4. Rate of return on common stock $$\qquad$$ in peak ML year $$-$$9.0% $$-$$45.5% $$\qquad$$ in peak and prior years $$-$$21.6% $$-$$61.1% 5. Altman z-score $$\qquad$$ at peak ML 2.35 1.94 $$\qquad$$ at post-peak trough 4.55 4.27 6. Return on assets (ROA) in year of $$\qquad$$ trough before peak ML 0.141 0.145 $$\qquad$$ peak ML 0.092 0.075 $$\qquad$$ trough after peak ML 0.145 0.139 7. Number of firms 1,407 2,466 8. Percentage of firms 36.3% 63.7% Movement to peak ML Median value of Proactive Other 1. Market leverage (ML) at peak 0.489 0.606 2. ML at post-peak trough 0.028 0.037 3. Percentage change in debt in peak ML year 93.4% $$-$$0.5% 4. Rate of return on common stock $$\qquad$$ in peak ML year $$-$$9.0% $$-$$45.5% $$\qquad$$ in peak and prior years $$-$$21.6% $$-$$61.1% 5. Altman z-score $$\qquad$$ at peak ML 2.35 1.94 $$\qquad$$ at post-peak trough 4.55 4.27 6. Return on assets (ROA) in year of $$\qquad$$ trough before peak ML 0.141 0.145 $$\qquad$$ peak ML 0.092 0.075 $$\qquad$$ trough after peak ML 0.145 0.139 7. Number of firms 1,407 2,466 8. Percentage of firms 36.3% 63.7% Market leverage (ML) is the ratio of the book value of debt to the sum of the book value of debt and the market value of equity. Peak leverage is the maximum value of ML over a firm’s entire time in the sample. The trough after peak is the lowest value of a firm’s ML that comes subsequent to its peak. The first column of the table examines firms that proactively increased ML in the year they attained peak ML (based on the conditions that Denis and McKeon (2012) use to identify large proactive ML increases). The second column examines all other firms, that is, all that do not satisfy the Denis and McKeon conditions. The table analyzes 3,873 of the 4,476 firms with at least five years of post-peak data. This subsample excludes 540 firms with peak ML in their first year on Compustat because the Denis and McKeon conditions require knowledge of the magnitude of leverage changes in the year in question and, for these firms, we do not have values for beginning-of-period ML. The subsample also excludes 63 firms with fiscal year end changes in the year of peak ML because of the difficulties of comparing ML changes calculated over heterogeneous intervals. Altman z-scores below 1.81 are generally interpreted as indicating a firm in distress that is a likely candidate for bankruptcy. Table 9 Deleveraging episodes: Sample partitioned by whether firms proactively move to peak market leverage Movement to peak ML Median value of Proactive Other 1. Market leverage (ML) at peak 0.489 0.606 2. ML at post-peak trough 0.028 0.037 3. Percentage change in debt in peak ML year 93.4% $$-$$0.5% 4. Rate of return on common stock $$\qquad$$ in peak ML year $$-$$9.0% $$-$$45.5% $$\qquad$$ in peak and prior years $$-$$21.6% $$-$$61.1% 5. Altman z-score $$\qquad$$ at peak ML 2.35 1.94 $$\qquad$$ at post-peak trough 4.55 4.27 6. Return on assets (ROA) in year of $$\qquad$$ trough before peak ML 0.141 0.145 $$\qquad$$ peak ML 0.092 0.075 $$\qquad$$ trough after peak ML 0.145 0.139 7. Number of firms 1,407 2,466 8. Percentage of firms 36.3% 63.7% Movement to peak ML Median value of Proactive Other 1. Market leverage (ML) at peak 0.489 0.606 2. ML at post-peak trough 0.028 0.037 3. Percentage change in debt in peak ML year 93.4% $$-$$0.5% 4. Rate of return on common stock $$\qquad$$ in peak ML year $$-$$9.0% $$-$$45.5% $$\qquad$$ in peak and prior years $$-$$21.6% $$-$$61.1% 5. Altman z-score $$\qquad$$ at peak ML 2.35 1.94 $$\qquad$$ at post-peak trough 4.55 4.27 6. Return on assets (ROA) in year of $$\qquad$$ trough before peak ML 0.141 0.145 $$\qquad$$ peak ML 0.092 0.075 $$\qquad$$ trough after peak ML 0.145 0.139 7. Number of firms 1,407 2,466 8. Percentage of firms 36.3% 63.7% Market leverage (ML) is the ratio of the book value of debt to the sum of the book value of debt and the market value of equity. Peak leverage is the maximum value of ML over a firm’s entire time in the sample. The trough after peak is the lowest value of a firm’s ML that comes subsequent to its peak. The first column of the table examines firms that proactively increased ML in the year they attained peak ML (based on the conditions that Denis and McKeon (2012) use to identify large proactive ML increases). The second column examines all other firms, that is, all that do not satisfy the Denis and McKeon conditions. The table analyzes 3,873 of the 4,476 firms with at least five years of post-peak data. This subsample excludes 540 firms with peak ML in their first year on Compustat because the Denis and McKeon conditions require knowledge of the magnitude of leverage changes in the year in question and, for these firms, we do not have values for beginning-of-period ML. The subsample also excludes 63 firms with fiscal year end changes in the year of peak ML because of the difficulties of comparing ML changes calculated over heterogeneous intervals. Altman z-scores below 1.81 are generally interpreted as indicating a firm in distress that is a likely candidate for bankruptcy. Table 9 indicates that firms that proactively move to peak ML are generally healthier than the sample firms that do not meet the Denis and McKeon conditions. Specifically, firms that proactively increase ML tend to attain lower peak levels of ML (row 1), with fewer signs of trouble when at peak (row 5) and a higher return on assets at peak (row 6). They also tend to increase debt sharply in the peak ML year, while the other firms decrease their debt slightly (row 3) while seeing their equity values fall sharply (row 4). Despite these differences, the typical deleveraging outcomes are quite similar, with both groups having the median firm attain near-zero ML at the post-peak trough. With one exception, the same qualitative relationships hold for firms with between one and four years of post-peak data, as reported in Table IA2 of the Internet Appendix. The exception concerns attenuated deleveraging, as discussed in Sections 4.2 and 4.3: median ML at the post-peak trough is higher for firms with one to four years of data than it is for firms with at least five years. As elaborated in Section 7, the proactive moves to peak for the many sample firms that meet the Denis and McKeon conditions are consistent with flexibility-based theories of capital structure, but not with traditional trade-off theories. The reason is that such cases entail managerial decisions to take on debt and move markedly above the firm’s estimated target leverage ratio (see footnote 6). Proactive movements away from target should never occur according to traditional trade-off theories, while flexibility-based theories are fully consistent with such movements. 4.5 The path to trough ML: Asset contraction versus expansion The path to trough ML is markedly different for firms that expand substantially relative to firms that shrink substantially while deleveraging, with retained earnings representing a key distinguishing feature. Earnings retention is typically large at firms that grow a lot. Firms that shrink a lot tend to be in serious trouble both at peak ML and at the later trough, with deleveraging typically occurring despite negative retained earnings. Both sets of firms tend to issue large amounts of stock while deleveraging. Table 10 analyzes the 9.7% of firms whose total assets contract by half or more as they go from peak ML to trough (first column) and the 19.1% of firms whose assets at least double (second column). For the median firm in both samples, ML at the post-peak trough is close to zero (row 4, Table 10), but the paths up to peak and then down to trough differ greatly. Table 10 Asset contraction versus expansion during deleveraging episodes Total assets during deleveraging halved doubled 1. Median percentage change in total assets (TA) from ML peak to trough $$-$$69.1% 247.1% 2. Median percentage change in TA from prior trough to peak 7.7% 37.6% 3. Median ML at peak 0.627 0.451 4. Median ML at post-peak trough 0.012 0.016 5. Percentage of firms that proactively increase ML in peak year 23.7% 70.3% 6. Median Altman z-score at peak ML 0.21 2.39 7. Median Altman z-score at post-peak trough $$-$$1.65 5.63 8. Median return on assets (ROA) in peak ML year $$-$$0.047 0.094 9. Median return on assets (ROA) in year of post-peak trough $$-$$0.196 0.151 10. Median percentage of years with losses during deleveraging episode 100.0% 0.0% 11. Median retained earnings $$\div$$ market value at peak ML $$-$$0.313 0.844 12. Median retained earnings $$\div$$ total assets at peak ML $$-$$0.290 0.747 13. Median percentage change in debt from peak to trough $$-$$98.7% $$-$$54.6% 14. Percentage of firms that are net issuers of stock during deleveraging 69.7% 88.4% 15. Median percentage of shares issued during deleveraging episode 18.1% 51.0% 16. Number of firms 957 1,883 17. Percentage of full sample of 9,866 firms 9.7% 19.1% Total assets during deleveraging halved doubled 1. Median percentage change in total assets (TA) from ML peak to trough $$-$$69.1% 247.1% 2. Median percentage change in TA from prior trough to peak 7.7% 37.6% 3. Median ML at peak 0.627 0.451 4. Median ML at post-peak trough 0.012 0.016 5. Percentage of firms that proactively increase ML in peak year 23.7% 70.3% 6. Median Altman z-score at peak ML 0.21 2.39 7. Median Altman z-score at post-peak trough $$-$$1.65 5.63 8. Median return on assets (ROA) in peak ML year $$-$$0.047 0.094 9. Median return on assets (ROA) in year of post-peak trough $$-$$0.196 0.151 10. Median percentage of years with losses during deleveraging episode 100.0% 0.0% 11. Median retained earnings $$\div$$ market value at peak ML $$-$$0.313 0.844 12. Median retained earnings $$\div$$ total assets at peak ML $$-$$0.290 0.747 13. Median percentage change in debt from peak to trough $$-$$98.7% $$-$$54.6% 14. Percentage of firms that are net issuers of stock during deleveraging 69.7% 88.4% 15. Median percentage of shares issued during deleveraging episode 18.1% 51.0% 16. Number of firms 957 1,883 17. Percentage of full sample of 9,866 firms 9.7% 19.1% This table compares firms whose total assets halved (contracted by 50% or more) while moving from peak ML to trough with firms whose total assets doubled (increased by 100% or more). Total assets (TA) is denominated in book value terms. Market leverage (ML) is the ratio of the book value of debt to the sum of the book value of debt and the market value of equity. Peak leverage is the maximum value of ML over a firm’s entire time in the sample. These data are for all 9,866 firms with at least one year of post-peak data on Compustat. The trough before (after) peak is the lowest value of a firm’s ML that comes prior to (subsequent to) its peak. Altman z-scores below 1.81 are generally interpreted as indicating a firm in distress and that is a likely candidate for bankruptcy. z-scores above 2.99 are generally viewed as indicating a firm that is safe, and z-scores between 1.81 and 2.99 are viewed as indicating that it is difficult to tell if the firm will experience distress. Row 5 reports the percentage of sample firms whose increase in ML in the peak year satisfies the sampling conditions in Denis and McKeon (2012) for a proactive increase in leverage. In row 10, the loss incidence is based on earnings before extraordinary items. Table 10 Asset contraction versus expansion during deleveraging episodes Total assets during deleveraging halved doubled 1. Median percentage change in total assets (TA) from ML peak to trough $$-$$69.1% 247.1% 2. Median percentage change in TA from prior trough to peak 7.7% 37.6% 3. Median ML at peak 0.627 0.451 4. Median ML at post-peak trough 0.012 0.016 5. Percentage of firms that proactively increase ML in peak year 23.7% 70.3% 6. Median Altman z-score at peak ML 0.21 2.39 7. Median Altman z-score at post-peak trough $$-$$1.65 5.63 8. Median return on assets (ROA) in peak ML year $$-$$0.047 0.094 9. Median return on assets (ROA) in year of post-peak trough $$-$$0.196 0.151 10. Median percentage of years with losses during deleveraging episode 100.0% 0.0% 11. Median retained earnings $$\div$$ market value at peak ML $$-$$0.313 0.844 12. Median retained earnings $$\div$$ total assets at peak ML $$-$$0.290 0.747 13. Median percentage change in debt from peak to trough $$-$$98.7% $$-$$54.6% 14. Percentage of firms that are net issuers of stock during deleveraging 69.7% 88.4% 15. Median percentage of shares issued during deleveraging episode 18.1% 51.0% 16. Number of firms 957 1,883 17. Percentage of full sample of 9,866 firms 9.7% 19.1% Total assets during deleveraging halved doubled 1. Median percentage change in total assets (TA) from ML peak to trough $$-$$69.1% 247.1% 2. Median percentage change in TA from prior trough to peak 7.7% 37.6% 3. Median ML at peak 0.627 0.451 4. Median ML at post-peak trough 0.012 0.016 5. Percentage of firms that proactively increase ML in peak year 23.7% 70.3% 6. Median Altman z-score at peak ML 0.21 2.39 7. Median Altman z-score at post-peak trough $$-$$1.65 5.63 8. Median return on assets (ROA) in peak ML year $$-$$0.047 0.094 9. Median return on assets (ROA) in year of post-peak trough $$-$$0.196 0.151 10. Median percentage of years with losses during deleveraging episode 100.0% 0.0% 11. Median retained earnings $$\div$$ market value at peak ML $$-$$0.313 0.844 12. Median retained earnings $$\div$$ total assets at peak ML $$-$$0.290 0.747 13. Median percentage change in debt from peak to trough $$-$$98.7% $$-$$54.6% 14. Percentage of firms that are net issuers of stock during deleveraging 69.7% 88.4% 15. Median percentage of shares issued during deleveraging episode 18.1% 51.0% 16. Number of firms 957 1,883 17. Percentage of full sample of 9,866 firms 9.7% 19.1% This table compares firms whose total assets halved (contracted by 50% or more) while moving from peak ML to trough with firms whose total assets doubled (increased by 100% or more). Total assets (TA) is denominated in book value terms. Market leverage (ML) is the ratio of the book value of debt to the sum of the book value of debt and the market value of equity. Peak leverage is the maximum value of ML over a firm’s entire time in the sample. These data are for all 9,866 firms with at least one year of post-peak data on Compustat. The trough before (after) peak is the lowest value of a firm’s ML that comes prior to (subsequent to) its peak. Altman z-scores below 1.81 are generally interpreted as indicating a firm in distress and that is a likely candidate for bankruptcy. z-scores above 2.99 are generally viewed as indicating a firm that is safe, and z-scores between 1.81 and 2.99 are viewed as indicating that it is difficult to tell if the firm will experience distress. Row 5 reports the percentage of sample firms whose increase in ML in the peak year satisfies the sampling conditions in Denis and McKeon (2012) for a proactive increase in leverage. In row 10, the loss incidence is based on earnings before extraordinary items. Median ML at peak is much higher among firms that contract by half than for firms that double in size (row 3, Table 10). Firms that shrink markedly after peak ML face much more serious trouble at peak and their trouble worsens despite their deleveraging to near-zero ML (rows 6 and 7). Most of these firms remain in serious trouble and the explanation is straightforward: Their sharp asset shrinkage reflects poor operating performance in the peak year and thereafter (rows 8 to 10), and so they typically generate no retained earnings to foster deleveraging (rows 11 and 12). They accordingly attempt to stay ahead of their continuing troubles by aggressively paying down debt (row 13) and issuing shares to bolster equity (rows 14 and 15). Among firms whose assets double or more in size after peak ML, high earnings (rows 8 and 9, Table 10) provide the basis for abundant earnings retention (rows 11 and 12) during their deleveraging episodes. Large earnings retention after peak ML, in turn, helps bring ML closer to zero even though the typical firm in this group repays only about half the debt it had outstanding at peak (row 13). Over the period leading up to peak ML, these firms expand by much larger amounts than firms whose assets shrink dramatically in the post-peak period (row 2, Table 10). It thus makes sense that managers of most of these aggressively expanding firms proactively lever up to peak ML (row 5). These firms’ trough-peak-trough pattern of asset growth and leverage has an intuitively plausible interpretation: Firms that are expanding a lot tend to see ML increase because, consistent with flexibility-based theories of capital structure, they often use debt to fund their growth and their subsequent deleveraging in part reflects material retained earnings that help fund expansion. 5. Cross-Firm Variation in Deleveraging Outcomes The empirical capital structure literature identifies industry leverage and a handful of firm-specific characteristics as factors that help explain leverage decisions. The latter variables include a firm’s current profitability, size, asset tangibility, market-to-book ratio, and whether or not it pays dividends (Rajan and Zingales 1995; Frank and Goyal 2009). The regressions in this section provide evidence of the extent to which cross-firm variation in deleveraging outcomes—ML at the post-peak trough—can be explained by traditional leverage determinants from the literature and by a number of other factors that are specific to firms’ deleveraging episodes. We find that firm leverage is both highly path dependent and closely linked to the restoration of financial flexibility in terms of the amount of cash the firm has accumulated as of the post-peak trough. For explaining whether ML at the outcome of deleveraging is relatively high or low, the key path-specific factors are knowledge of ML at the peak, ML at the prior trough, and whether the firm has had just a short time to work ML back down, for example, due to distress-related delisting. Traditional leverage determinants are statistically significant, but their stand-alone ability to explain the variation in ML at the post-peak trough falls well short of the explanatory power of this simple model. The overall explanatory power increases modestly when the simple model is expanded to include the traditional variables and factors such as whether peak ML was attained proactively and the extent of asset growth (or shrinkage) during the deleveraging. In Table 11’s regressions, the dependent variable is ML at the post-peak trough and the right-hand-side variables are traditional determinants from the literature evaluated when the firm is at its post-peak trough (rows 1 to 6); variables known as of peak: ML at the peak and ML at the trough before peak (rows 7 and 8); and variables known by the post-peak trough: Cash/TA at the trough as well as indicator variables that identify firms with four or fewer years of post-peak data and firms delisted due to financial distress or merger within four years of attaining peak (rows 9 to 12). Table 11 Cross-firm variation in market leverage (ML) at the ML trough after peak (1) (2) (3) (4) (5) (6) (7) (8) Traditional factors at post-peak trough: 1. Industry ML at trough after peak 0.495 0.147 (8.91) (4.45) 2. Profitability (ROA) at trough –0.066 –0.033 (–2.40) (–1.67) 3 Size at trough 0.009 0.010 (2.55) (7.25) 4. Market to book at trough –0.017 –0.005 (–5.34) (–3.74) 5. Asset tangibility at trough 0.213 0.066 (3.43) (2.13) 6. Paid dividends in trough year –0.080 –0.010 (–6.56) (–1.18) Known as of peak: 7. ML at trough before peak 0.525 0.181 0.172 0.145 (12.54) (6.62) (8.54) (9.35) 8. ML at peak 0.434 0.417 0.395 (20.00) (19.43) (19.60) Known by post-peak trough: 9. $$<$$ 5 years of post-peak data 0.132 0.123 0.152 0.150 (10.68) (10.19) (9.61) (9.18) 10. Distress delist $$<$$ 5 years after peak 0.077 0.074 0.041 0.053 (4.81) (5.30) (2.56) (3.59) 11. Merger delist $$<$$ 5 years after peak 0.017 –0.011 0.038 0.028 (0.11) (–0.99) (2.73) (1.98) 12. Cash/Total assets at trough –0.432 –0.271 –0.161 –0.105 (–9.39) (–7.04) (–5.96) (–3.68) 13. Number of firms 8,328 8,328 8,328 8,328 8,328 8,328 8,328 8,328 14. Adjusted R$$^{\mathrm{2}}$$ without FEs 19% 15% 36% 13% 15% 34% 53% 55% 15. Adjusted R$$^{\mathrm{2}}$$ with FEs 26% 27% 50% 22% 26% 37% 57% 59% (1) (2) (3) (4) (5) (6) (7) (8) Traditional factors at post-peak trough: 1. Industry ML at trough after peak 0.495 0.147 (8.91) (4.45) 2. Profitability (ROA) at trough –0.066 –0.033 (–2.40) (–1.67) 3 Size at trough 0.009 0.010 (2.55) (7.25) 4. Market to book at trough –0.017 –0.005 (–5.34) (–3.74) 5. Asset tangibility at trough 0.213 0.066 (3.43) (2.13) 6. Paid dividends in trough year –0.080 –0.010 (–6.56) (–1.18) Known as of peak: 7. ML at trough before peak 0.525 0.181 0.172 0.145 (12.54) (6.62) (8.54) (9.35) 8. ML at peak 0.434 0.417 0.395 (20.00) (19.43) (19.60) Known by post-peak trough: 9. $$<$$ 5 years of post-peak data 0.132 0.123 0.152 0.150 (10.68) (10.19) (9.61) (9.18) 10. Distress delist $$<$$ 5 years after peak 0.077 0.074 0.041 0.053 (4.81) (5.30) (2.56) (3.59) 11. Merger delist $$<$$ 5 years after peak 0.017 –0.011 0.038 0.028 (0.11) (–0.99) (2.73) (1.98) 12. Cash/Total assets at trough –0.432 –0.271 –0.161 –0.105 (–9.39) (–7.04) (–5.96) (–3.68) 13. Number of firms 8,328 8,328 8,328 8,328 8,328 8,328 8,328 8,328 14. Adjusted R$$^{\mathrm{2}}$$ without FEs 19% 15% 36% 13% 15% 34% 53% 55% 15. Adjusted R$$^{\mathrm{2}}$$ with FEs 26% 27% 50% 22% 26% 37% 57% 59% Market leverage (ML) is the ratio of book debt to the sum of book debt and the market value of equity. The left-hand-side variable is ML at the post-peak trough. Rows 1 to 6 have traditional leverage determinants as in Rajan and Zingales (1995) and Frank and Goyal (2009), with all variables evaluated at the trough after peak ML. Row 1 employs the industry median ML ratio. Rows 6 and 9 to 11 have zero-one indicator variables that equal one if a firm has the specified characteristic. Standard errors are clustered by peak year and industry, and $$t$$-statistics are in parentheses. In all cases, the reported coefficients are for models without fixed effects. Row 15 reports adjusted R$$^{\mathrm{2}}$$s for the indicated specification with fixed effects (FEs) for industry and years in the deleveraging episode. The reduction in sample size from 9,866 firms reflects missing data items on Compustat for some variables. The sample size is 8,328 firms in all cases because that is the largest number of observations (with nonmissing data items) for which we can keep the sample composition constant across models. Table 11 Cross-firm variation in market leverage (ML) at the ML trough after peak (1) (2) (3) (4) (5) (6) (7) (8) Traditional factors at post-peak trough: 1. Industry ML at trough after peak 0.495 0.147 (8.91) (4.45) 2. Profitability (ROA) at trough –0.066 –0.033 (–2.40) (–1.67) 3 Size at trough 0.009 0.010 (2.55) (7.25) 4. Market to book at trough –0.017 –0.005 (–5.34) (–3.74) 5. Asset tangibility at trough 0.213 0.066 (3.43) (2.13) 6. Paid dividends in trough year –0.080 –0.010 (–6.56) (–1.18) Known as of peak: 7. ML at trough before peak 0.525 0.181 0.172 0.145 (12.54) (6.62) (8.54) (9.35) 8. ML at peak 0.434 0.417 0.395 (20.00) (19.43) (19.60) Known by post-peak trough: 9. $$<$$ 5 years of post-peak data 0.132 0.123 0.152 0.150 (10.68) (10.19) (9.61) (9.18) 10. Distress delist $$<$$ 5 years after peak 0.077 0.074 0.041 0.053 (4.81) (5.30) (2.56) (3.59) 11. Merger delist $$<$$ 5 years after peak 0.017 –0.011 0.038 0.028 (0.11) (–0.99) (2.73) (1.98) 12. Cash/Total assets at trough –0.432 –0.271 –0.161 –0.105 (–9.39) (–7.04) (–5.96) (–3.68) 13. Number of firms 8,328 8,328 8,328 8,328 8,328 8,328 8,328 8,328 14. Adjusted R$$^{\mathrm{2}}$$ without FEs 19% 15% 36% 13% 15% 34% 53% 55% 15. Adjusted R$$^{\mathrm{2}}$$ with FEs 26% 27% 50% 22% 26% 37% 57% 59% (1) (2) (3) (4) (5) (6) (7) (8) Traditional factors at post-peak trough: 1. Industry ML at trough after peak 0.495 0.147 (8.91) (4.45) 2. Profitability (ROA) at trough –0.066 –0.033 (–2.40) (–1.67) 3 Size at trough 0.009 0.010 (2.55) (7.25) 4. Market to book at trough –0.017 –0.005 (–5.34) (–3.74) 5. Asset tangibility at trough 0.213 0.066 (3.43) (2.13) 6. Paid dividends in trough year –0.080 –0.010 (–6.56) (–1.18) Known as of peak: 7. ML at trough before peak 0.525 0.181 0.172 0.145 (12.54) (6.62) (8.54) (9.35) 8. ML at peak 0.434 0.417 0.395 (20.00) (19.43) (19.60) Known by post-peak trough: 9. $$<$$ 5 years of post-peak data 0.132 0.123 0.152 0.150 (10.68) (10.19) (9.61) (9.18) 10. Distress delist $$<$$ 5 years after peak 0.077 0.074 0.041 0.053 (4.81) (5.30) (2.56) (3.59) 11. Merger delist $$<$$ 5 years after peak 0.017 –0.011 0.038 0.028 (0.11) (–0.99) (2.73) (1.98) 12. Cash/Total assets at trough –0.432 –0.271 –0.161 –0.105 (–9.39) (–7.04) (–5.96) (–3.68) 13. Number of firms 8,328 8,328 8,328 8,328 8,328 8,328 8,328 8,328 14. Adjusted R$$^{\mathrm{2}}$$ without FEs 19% 15% 36% 13% 15% 34% 53% 55% 15. Adjusted R$$^{\mathrm{2}}$$ with FEs 26% 27% 50% 22% 26% 37% 57% 59% Market leverage (ML) is the ratio of book debt to the sum of book debt and the market value of equity. The left-hand-side variable is ML at the post-peak trough. Rows 1 to 6 have traditional leverage determinants as in Rajan and Zingales (1995) and Frank and Goyal (2009), with all variables evaluated at the trough after peak ML. Row 1 employs the industry median ML ratio. Rows 6 and 9 to 11 have zero-one indicator variables that equal one if a firm has the specified characteristic. Standard errors are clustered by peak year and industry, and $$t$$-statistics are in parentheses. In all cases, the reported coefficients are for models without fixed effects. Row 15 reports adjusted R$$^{\mathrm{2}}$$s for the indicated specification with fixed effects (FEs) for industry and years in the deleveraging episode. The reduction in sample size from 9,866 firms reflects missing data items on Compustat for some variables. The sample size is 8,328 firms in all cases because that is the largest number of observations (with nonmissing data items) for which we can keep the sample composition constant across models. When the traditional leverage determinants are treated as a stand-alone model of deleveraging, their coefficients are statistically significant and of the signs found in prior studies (column 1, rows 1 to 6, Table 11). They collectively explain 19% of the variation in ML at the post-peak trough and, when the regression model also includes industry and year fixed effects, the adjusted R$$^{2}$$ is 26% (rows 14 and 15). We find similar explanatory power for a basic model in which the only right-hand-side variable is ML at the trough before peak. The adjusted R$$^{2}$$s are 15% and 27%, respectively, for this model without and with fixed effects (rows 14 and 15, column 2, Table 11)). The coefficient on the pre-peak trough is positive and highly significant (row 7, column 2), and this finding is robust, a fact that will soon become apparent. It is tempting to view the latter finding as indicative of leverage rebalancing over long horizons to a target that is stationary or nearly so. However, such an explanation for our finding is implausible since so many firms spend years moving away from the previous trough with discretionary debt issuance. The targeting behavior implied by standard trade-off theories of capital structure would imply that firms takes actions to move toward their target or, at the very least, do not choose to move away from it. As Section 7 explains, our finding is more compatible with deleveraging to replenish flexibility, which most firms had in abundance at the pre-peak trough and which most also have in abundance at the post-peak trough. In Table 11, an indication of an important role for flexibility considerations is provided by regression (5) in which the only explanatory variable is the firm’s cash-balance ratio at the outcome of deleveraging. In this regression, greater flexibility in terms of higher Cash/TA indicates that the firm is also expected to have greater flexibility in terms of a lower ML ratio at the post-peak trough. This model has adjusted R$$^{2}$$s of 15% and 26% (rows 14 and 15), which are very close to the adjusted R$$^{2}$$s for the model with traditional variables. It is remarkable that simply having knowledge of a firm’s cash holdings—an element of flexibility that is complementary to the capacity to borrow—does as good a job explaining its deleveraging outcome as the set of traditionally specified variables that decades of research have found to be the strongest known determinants of leverage. Other regression models in Table 11 have considerably greater explanatory power than the traditional variables. For example, when ML at the peak is included as an explanatory variable along with ML at the pre-peak trough, the R$$^{2}$$s increase to 36% and 50% (rows 14 and 15, column 3). In other words, a simple model that includes only information known at the beginning of the deleveraging episode explains about twice the cross-firm variation in deleveraging outcomes as do models based solely on traditional leverage determinants measured at the end of the episode. The R$$^{2}$$s increase to 53% and 57% when the Table 11 regressions also take into account the cash-balances the firm has accumulated at the trough and the number of years of post-peak data a firm has on Compustat (column 7). As with the univariate specification, greater financial flexibility as measured by a higher Cash/TA at the outcome of deleveraging indicates greater flexibility in terms of lower leverage at the same point in time (row 12, column 7). The fact that the cash component of financial flexibility is a reliable predictor of the leverage component is consistent with the view that our firms are concerned with rebuilding flexibility generally, and not just with reducing ML to a low level. The still-simple model in column 7 indicates that having fewer than five years of data after peak implies that ML at the trough is typically 0.152 higher than the trough ML for firms with longer post-peak data periods (row 9). The ML troughs are typically higher by another 0.041 and 0.038, respectively, after taking into account whether a firm is delisted due to distress or merger soon after reaching peak (rows 10 and 11). These estimates together imply a total difference in deleveraging outcomes of around 0.200 for distress and merger delists over and above other sample firms. This large difference understates the extent of attenuation in deleveraging for these two types of delists because the regression sample excludes quite a few cases (per Section 4) in which firms are delisted due to distress or merger in the peak year. When the traditional leverage determinants are added to the column 7 model, the R$$^{2}$$s increase by only 2% to 55% and 59%, respectively (rows 14 and 15, column 8, Table 11). All traditional variables retain the signs usually reported in prior studies, but there are some indications of diminished significance levels (compare columns 1 and 8, rows 1 to 6). Table 12 expands the analysis of deleveraging outcomes to include several other factors that are known as of peak ML (rows 8 to 12) as well as factors that become known between peak and trough (rows 17 and 18).7 Inclusion of this broader set of factors does not change the main carry-away about the key sources of explanatory power for deleveraging outcomes, as the all-inclusive models in columns 4 and 5 of Table 12 have R$$^{2}$$s that are only slightly above the R$$^{2}$$ for the model in column 8 of Table 11. Table 12 Deleveraging outcomes: Impact of financial distress, proactive leverage increases, and recession timing (1) (2) (3) (4) (5) Traditional factors at post-peak trough: 1. Industry ML at trough after peak 0.494 0.151 0.143 0.213 (9.63) (4.52) (4.78) (5.63) 2. Profitability (ROA) at trough –0.063 –0.034 –0.053 –0.050 (–2.33) (–1.84) (–2.72) (–2.47) 3 Size at trough 0.008 0.010 0.010 0.014 (2.18) (6.56) (5.87) (7.01) 4. Market to book at trough –0.017 –0.004 –0.003 –0.003 (–5.07) (–3.71) (–3.09) (–2.34) 5. Asset tangibility at trough 0.208 0.059 0.056 0.094 (3.73) (2.18) (2.27) (6.18) 6. Paid dividends in trough year –0.075 –0.006 –0.005 0.003 (–5.98) (–0.86) (–0.67) (0.41) Known as of peak: 7. ML at trough before peak 0.177 0.146 0.153 0.151 (8.31) (8.97) (10.40) (8.67) 8. ML at peak x distressed at peak 0.413 0.418 0.425 0.448 (18.98) (19.10) (19.77) (18.20) 9. ML at peak x not distressed at peak 0.417 0.390 0.394 0.428 (17.39) (14.66) (15.19) (15.66) 10. Financially distressed at peak –0.018 –0.018 –0.013 (–1.99) (–2.12) (–1.27) 11. Proactive increase to peak ML 0.016 0.016 0.014 (3.97) (4.54) (3.16) 12. Peak ML during recession –0.013 –0.012 (–1.33) (–1.39) Known by post-peak trough: 13. $$<$$ 5 years of post-peak data 0.150 0.150 0.132 0.106 (9.14) (10.00) (9.82) (10.76) 14. Distress delist $$<$$ 5 years after peak 0.042 0.051 0.056 0.052 (2.55) (3.62) (4.14) (5.01) 15. Merger delist $$<$$ 5 years after peak 0.035 0.023 0.022 0.012 (2.46) (1.71) (1.64) (1.49) 16. Cash/Total assets at trough –0.156 –0.102 –0.089 –0.092 (–6.17) (–3.83) (–3.78) (–4.30) 17. Assets halved from peak to trough –0.077 –0.065 (–7.76) (–5.50) 18. Assets doubled from peak to trough –0.062 –0.019 (–8.16) (–2.90) 19. Number of firms 7,125 7,125 7,125 7,125 7,125 20. Adjusted R$$^{\mathrm{2}}$$ without FEs 19% 53% 55% 57% – 21. Adjusted R$$^{\mathrm{2}}$$ with FEs 26% 57% – – 60% (1) (2) (3) (4) (5) Traditional factors at post-peak trough: 1. Industry ML at trough after peak 0.494 0.151 0.143 0.213 (9.63) (4.52) (4.78) (5.63) 2. Profitability (ROA) at trough –0.063 –0.034 –0.053 –0.050 (–2.33) (–1.84) (–2.72) (–2.47) 3 Size at trough 0.008 0.010 0.010 0.014 (2.18) (6.56) (5.87) (7.01) 4. Market to book at trough –0.017 –0.004 –0.003 –0.003 (–5.07) (–3.71) (–3.09) (–2.34) 5. Asset tangibility at trough 0.208 0.059 0.056 0.094 (3.73) (2.18) (2.27) (6.18) 6. Paid dividends in trough year –0.075 –0.006 –0.005 0.003 (–5.98) (–0.86) (–0.67) (0.41) Known as of peak: 7. ML at trough before peak 0.177 0.146 0.153 0.151 (8.31) (8.97) (10.40) (8.67) 8. ML at peak x distressed at peak 0.413 0.418 0.425 0.448 (18.98) (19.10) (19.77) (18.20) 9. ML at peak x not distressed at peak 0.417 0.390 0.394 0.428 (17.39) (14.66) (15.19) (15.66) 10. Financially distressed at peak –0.018 –0.018 –0.013 (–1.99) (–2.12) (–1.27) 11. Proactive increase to peak ML 0.016 0.016 0.014 (3.97) (4.54) (3.16) 12. Peak ML during recession –0.013 –0.012 (–1.33) (–1.39) Known by post-peak trough: 13. $$<$$ 5 years of post-peak data 0.150 0.150 0.132 0.106 (9.14) (10.00) (9.82) (10.76) 14. Distress delist $$<$$ 5 years after peak 0.042 0.051 0.056 0.052 (2.55) (3.62) (4.14) (5.01) 15. Merger delist $$<$$ 5 years after peak 0.035 0.023 0.022 0.012 (2.46) (1.71) (1.64) (1.49) 16. Cash/Total assets at trough –0.156 –0.102 –0.089 –0.092 (–6.17) (–3.83) (–3.78) (–4.30) 17. Assets halved from peak to trough –0.077 –0.065 (–7.76) (–5.50) 18. Assets doubled from peak to trough –0.062 –0.019 (–8.16) (–2.90) 19. Number of firms 7,125 7,125 7,125 7,125 7,125 20. Adjusted R$$^{\mathrm{2}}$$ without FEs 19% 53% 55% 57% – 21. Adjusted R$$^{\mathrm{2}}$$ with FEs 26% 57% – – 60% Market leverage (ML) is the ratio of book debt to the sum of book debt and the market value of equity. In all regressions, the left-hand-side variable is ML at the post-peak trough. Rows 1 to 7 and 13 to 15 are as defined in Table 10. Rows 1 to 6 have traditional leverage determinants. Row 8 has ML at the peak for firms with Altman z-scores at peak $$<$$ 1.81. Row 9 has ML at the peak for firms with z-scores $$\geqslant$$ 1.81. Rows 6, 10 to 15, 17, and 18 have zero-one indicator variables that equal one if the firm has a particular specified characteristic. For row 10, distressed firms are those with z-scores $$<$$ 1.81 at peak ML. The row 11 variable equals one if ML at peak satisfies the Denis and McKeon (2012) conditions for a proactive increase. The row 12 variable equals one if the ML peak is within six months of an NBER-defined recession. Row 21 reports adjusted R$$^{\mathrm{2}}$$s for the same column model with fixed effects (FEs) for industry and years in the deleveraging episode. Standard errors are clustered by peak year and industry, and $$t$$-statistics are in parentheses. The sample size is 7,125 because that is the largest number of firms (with nonmissing data items) for which we can keep the sample composition fixed across models. Table 12 Deleveraging outcomes: Impact of financial distress, proactive leverage increases, and recession timing (1) (2) (3) (4) (5) Traditional factors at post-peak trough: 1. Industry ML at trough after peak 0.494 0.151 0.143 0.213 (9.63) (4.52) (4.78) (5.63) 2. Profitability (ROA) at trough –0.063 –0.034 –0.053 –0.050 (–2.33) (–1.84) (–2.72) (–2.47) 3 Size at trough 0.008 0.010 0.010 0.014 (2.18) (6.56) (5.87) (7.01) 4. Market to book at trough –0.017 –0.004 –0.003 –0.003 (–5.07) (–3.71) (–3.09) (–2.34) 5. Asset tangibility at trough 0.208 0.059 0.056 0.094 (3.73) (2.18) (2.27) (6.18) 6. Paid dividends in trough year –0.075 –0.006 –0.005 0.003 (–5.98) (–0.86) (–0.67) (0.41) Known as of peak: 7. ML at trough before peak 0.177 0.146 0.153 0.151 (8.31) (8.97) (10.40) (8.67) 8. ML at peak x distressed at peak 0.413 0.418 0.425 0.448 (18.98) (19.10) (19.77) (18.20) 9. ML at peak x not distressed at peak 0.417 0.390 0.394 0.428 (17.39) (14.66) (15.19) (15.66) 10. Financially distressed at peak –0.018 –0.018 –0.013 (–1.99) (–2.12) (–1.27) 11. Proactive increase to peak ML 0.016 0.016 0.014 (3.97) (4.54) (3.16) 12. Peak ML during recession –0.013 –0.012 (–1.33) (–1.39) Known by post-peak trough: 13. $$<$$ 5 years of post-peak data 0.150 0.150 0.132 0.106 (9.14) (10.00) (9.82) (10.76) 14. Distress delist $$<$$ 5 years after peak 0.042 0.051 0.056 0.052 (2.55) (3.62) (4.14) (5.01) 15. Merger delist $$<$$ 5 years after peak 0.035 0.023 0.022 0.012 (2.46) (1.71) (1.64) (1.49) 16. Cash/Total assets at trough –0.156 –0.102 –0.089 –0.092 (–6.17) (–3.83) (–3.78) (–4.30) 17. Assets halved from peak to trough –0.077 –0.065 (–7.76) (–5.50) 18. Assets doubled from peak to trough –0.062 –0.019 (–8.16) (–2.90) 19. Number of firms 7,125 7,125 7,125 7,125 7,125 20. Adjusted R$$^{\mathrm{2}}$$ without FEs 19% 53% 55% 57% – 21. Adjusted R$$^{\mathrm{2}}$$ with FEs 26% 57% – – 60% (1) (2) (3) (4) (5) Traditional factors at post-peak trough: 1. Industry ML at trough after peak 0.494 0.151 0.143 0.213 (9.63) (4.52) (4.78) (5.63) 2. Profitability (ROA) at trough –0.063 –0.034 –0.053 –0.050 (–2.33) (–1.84) (–2.72) (–2.47) 3 Size at trough 0.008 0.010 0.010 0.014 (2.18) (6.56) (5.87) (7.01) 4. Market to book at trough –0.017 –0.004 –0.003 –0.003 (–5.07) (–3.71) (–3.09) (–2.34) 5. Asset tangibility at trough 0.208 0.059 0.056 0.094 (3.73) (2.18) (2.27) (6.18) 6. Paid dividends in trough year –0.075 –0.006 –0.005 0.003 (–5.98) (–0.86) (–0.67) (0.41) Known as of peak: 7. ML at trough before peak 0.177 0.146 0.153 0.151 (8.31) (8.97) (10.40) (8.67) 8. ML at peak x distressed at peak 0.413 0.418 0.425 0.448 (18.98) (19.10) (19.77) (18.20) 9. ML at peak x not distressed at peak 0.417 0.390 0.394 0.428 (17.39) (14.66) (15.19) (15.66) 10. Financially distressed at peak –0.018 –0.018 –0.013 (–1.99) (–2.12) (–1.27) 11. Proactive increase to peak ML 0.016 0.016 0.014 (3.97) (4.54) (3.16) 12. Peak ML during recession –0.013 –0.012 (–1.33) (–1.39) Known by post-peak trough: 13. $$<$$ 5 years of post-peak data 0.150 0.150 0.132 0.106 (9.14) (10.00) (9.82) (10.76) 14. Distress delist $$<$$ 5 years after peak 0.042 0.051 0.056 0.052 (2.55) (3.62) (4.14) (5.01) 15. Merger delist $$<$$ 5 years after peak 0.035 0.023 0.022 0.012 (2.46) (1.71) (1.64) (1.49) 16. Cash/Total assets at trough –0.156 –0.102 –0.089 –0.092 (–6.17) (–3.83) (–3.78) (–4.30) 17. Assets halved from peak to trough –0.077 –0.065 (–7.76) (–5.50) 18. Assets doubled from peak to trough –0.062 –0.019 (–8.16) (–2.90) 19. Number of firms 7,125 7,125 7,125 7,125 7,125 20. Adjusted R$$^{\mathrm{2}}$$ without FEs 19% 53% 55% 57% – 21. Adjusted R$$^{\mathrm{2}}$$ with FEs 26% 57% – – 60% Market leverage (ML) is the ratio of book debt to the sum of book debt and the market value of equity. In all regressions, the left-hand-side variable is ML at the post-peak trough. Rows 1 to 7 and 13 to 15 are as defined in Table 10. Rows 1 to 6 have traditional leverage determinants. Row 8 has ML at the peak for firms with Altman z-scores at peak $$<$$ 1.81. Row 9 has ML at the peak for firms with z-scores $$\geqslant$$ 1.81. Rows 6, 10 to 15, 17, and 18 have zero-one indicator variables that equal one if the firm has a particular specified characteristic. For row 10, distressed firms are those with z-scores $$<$$ 1.81 at peak ML. The row 11 variable equals one if ML at peak satisfies the Denis and McKeon (2012) conditions for a proactive increase. The row 12 variable equals one if the ML peak is within six months of an NBER-defined recession. Row 21 reports adjusted R$$^{\mathrm{2}}$$s for the same column model with fixed effects (FEs) for industry and years in the deleveraging episode. Standard errors are clustered by peak year and industry, and $$t$$-statistics are in parentheses. The sample size is 7,125 because that is the largest number of firms (with nonmissing data items) for which we can keep the sample composition fixed across models. Table 12 has four other findings of interest. First, there are at best minor differences in deleveraging outcomes between firms that are distressed when they are at peak ML and those that are not. The right-hand-side variables now include ML at peak interacted with an indicator variable for firms that are distressed at peak (rows 8 and 9) as well as the indicator variable itself (row 10). These refinements have negligible effects on overall explanatory power. Moreover, the estimated distress indicator effect is small and slightly negative (row 10) rather than positive as one would expect if being distressed at peak implies systematically higher ML at the trough. A plausible reason is that the strong explanatory role of ML at the peak likely reflects the fact that financial distress is capitalized into a lower equity value at peak which, of course, translates to a higher ML ratio at that point in time. Second, firms that proactively lever up to peak tend to have ML ratios at the subsequent trough that are significantly higher in a statistical sense, but the typical difference is less than 0.020 and thus of minor economic significance (row 11). Third, the estimated outcome differential is of roughly the same minor magnitude for deleveraging episodes that begin during recessions (row 12, columns 3 and 4, Table 12). Fourth, large asset growth or contraction is associated with significantly lower ML at the post-peak trough (rows 17 and 18). While the point estimates of the asset-structure effects are reasonably large, inclusion of these variables increases the adjusted R$$^{2}$$ by only 2% (columns 3 and 4, row 20). Overall, then, while the refinements in Table 12 improve the ability to explain differences across firms in deleveraging outcomes, their incremental explanatory power is modest at best. The main carry-away from Table 11 thus remains relevant: The key to understanding the extent of cross-firm variation in deleveraging outcomes is information about path dependency (in terms of a firm’s ML ratios at peak and prior trough and whether it has had only a few years to deleverage) and the extent to which a firm has rebuilt financial flexibility through cash accumulation. 6. Differences from Prior Studies and Robustness Checks This section describes the key deleveraging findings of prior studies and explains why their findings on trends in cross-sectional average leverage ratios understate the extent of firm-level deleveraging by nonfinancial firms with high and/or recently increased leverage. It also presents robustness analysis that indicates that our main findings continue to hold qualitatively when we examine deleveraging after large spikes in leverage (with no requirement that firms begin deleveraging from all-time peak). 6.1 Prior studies examine event-time trends in cross-sectional average leverage ratios Prior studies report that deleveraging is typically much more muted than we find, and they report no indication of pervasive deleveraging to near-zero leverage. Among prior studies, the largest deleveraging is reported by Denis and McKeon (2012, figure 2 and table 4), who analyze the evolution of ML after large proactive increases in ML that, on average, move firms 0.270 above their estimated target ratios. They find that, over the seven years after these ML increases, cross-firm average ML declines from almost 0.550 to just over 0.400. The moderate ML decline of 0.133 occurs gradually (i.e., average ML drifts down by less than 0.020 per year). This decline offsets slightly more than 50% of the average ML spike of 0.240 from seven years earlier, and leaves average ML 0.122 above the average estimated target. Harford, Klasa, and Walcott (2009, table 6) report an average decline of around 0.060 in excess leverage (actual leverage minus an estimated target ratio) over the five years following cash-financed acquisitions. Their sample does not require that leverage be “high” in any sense at the time of the cash-based acquisition. However, it is similar to the study by Denis and McKeon in that both examine the time path of average leverage ratios following managerial decisions to increase leverage; that is, exogenous leverage increases are excluded from both samples by design. Leary and Roberts (2005, figure 4, Panel C) report deleveraging estimates close in size to those in Harford, Klasa, and Walcott (2009). They analyze event-time averages of the difference in ML between firms that experience a large negative stock return and those that do not. The event-time average (of differences in ML) increases by about 0.100 or so in the year of the negative return and then, over the next five years, declines by about 0.060 or 0.080 (depending on the subsample). Finally, Lemmon, Roberts, and Zender (LRZ, 2008, figures 1 and 2) and DeAngelo and Roll (2015, p. 392) analyze average leverage ratios over 20-year periods for samples of firms with leverage initially in the top quartile of the cross-section. Both studies find that, although average leverage for the top quartile declines over time, it remains high in absolute terms and well above the average leverage of the other three quartiles. For example, LRZ (figure 1, panel C) find that, for firms in the top quartile, the sample average ML declines from a bit more than 0.600 in the sample-formation year to slightly below 0.500 two decades later. 6.2 Trends in event-time sample averages understate the typical scale of deleveraging The approach of these five studies—averaging first across firms at a given point in event time, and then examining the sequence of resultant sample average leverage ratios—yields information about the leverage of the typical firm at each event-time node after sample formation. DeAngelo and Roll (2015, p. 392) point out that trends in cross-sectional average leverage ratios can be misleading because averaging masks the empirically large time-series volatility in the leverage of most individual firms. For our study, the problem with this approach is not the masking of volatility. Rather, the problem is that it yields a downward-biased estimate of the size of the typical firm’s deleveraging. The bias arises from the joint impact of two empirically important features of the time path of leverage. First, the length of deleveraging episodes differs across firms. Second, after deleveraging to the post-peak trough, firms can only keep leverage constant or increase it.8 In a given event year after sample formation, the leverage cross-section will include some firms that are reducing their leverage and others that are now increasing their leverage after having reduced it. Inclusion of the latter firms pulls the cross-sectional event-time sample average up, thereby giving a misleading (understated) impression of the typical scale of deleveraging within the sample being analyzed. As Figure 6 shows, firms that deleverage relatively quickly to trough leverage and then increase leverage will inflate the cross-firm average leverage ratio during the event years that other firms are still deleveraging. This effect does not disappear over time because, with heterogeneity across firms in deleveraging, there is no future date at which all firms have fully deleveraged and remain at a new and stable level of leverage. There is nothing wrong with examining sample means or medians at the beginning or end of deleveraging episodes. The bias arises from examining changes in cross-firm event-time averages when episode length differs across firms. Figure 6 View largeDownload slide Why changes in event-time sample average leverage ratios give downward-biased estimates of the magnitude of deleveraging Prior studies measure the magnitude of deleveraging by the change in cross-sectional average (mean or median) leverage ratios calculated in event time. The figure above illustrates why this approach yields downward-biased measures of deleveraging. In this simple example, firms 1 and 2 start at the same peak leverage, L$$_{\rm peak}$$, and both deleverage to a zero-debt capital structure, but at different rates. Both firms take on new debt after reaching the trough at zero leverage. Since both firms reduce leverage from L$$_{\rm peak}$$ to zero, the true mean (and median) magnitude of the deleveraging in these two episodes is L$$_{\rm peak}$$. However, if one plots the average leverage ratio in event time relative to peak, leverage bottoms out at L$$_{\rm min}$$. The measured deleveraging is L$$_{\rm peak}$$–L$$_{\rm min}$$, which is well below the true amount, L$$_{\rm peak}$$. This attenuation reflects (1) cyclicality in leverage (with both firms taking on debt after their deleveraging episodes end) and (2) heterogeneity in the rate at which firms deleverage. In this example, the height to the minimum point on the event-time sample average plot represents the bias inherent in using the event-time mean or median leverage to gauge the magnitude of deleveraging by the typical firm. In other examples, the Sample Average plot will not be flat after Firm 1’s deleveraging episode ends. In general, the slope of the Sample Average plot depends on the rate of deleveraging at Firm 2 versus the rate that leverage increases at Firm 1 after the end of its deleveraging episode. In all cases, as long as some firms are increasing leverage while others are reducing deleverage, which is virtually always true in the real data, the event-time sample mean and median leverage ratios will yield downward-biased measures of the magnitude of deleveraging by the typical firm. Figure 6 View largeDownload slide Why changes in event-time sample average leverage ratios give downward-biased estimates of the magnitude of deleveraging Prior studies measure the magnitude of deleveraging by the change in cross-sectional average (mean or median) leverage ratios calculated in event time. The figure above illustrates why this approach yields downward-biased measures of deleveraging. In this simple example, firms 1 and 2 start at the same peak leverage, L$$_{\rm peak}$$, and both deleverage to a zero-debt capital structure, but at different rates. Both firms take on new debt after reaching the trough at zero leverage. Since both firms reduce leverage from L$$_{\rm peak}$$ to zero, the true mean (and median) magnitude of the deleveraging in these two episodes is L$$_{\rm peak}$$. However, if one plots the average leverage ratio in event time relative to peak, leverage bottoms out at L$$_{\rm min}$$. The measured deleveraging is L$$_{\rm peak}$$–L$$_{\rm min}$$, which is well below the true amount, L$$_{\rm peak}$$. This attenuation reflects (1) cyclicality in leverage (with both firms taking on debt after their deleveraging episodes end) and (2) heterogeneity in the rate at which firms deleverage. In this example, the height to the minimum point on the event-time sample average plot represents the bias inherent in using the event-time mean or median leverage to gauge the magnitude of deleveraging by the typical firm. In other examples, the Sample Average plot will not be flat after Firm 1’s deleveraging episode ends. In general, the slope of the Sample Average plot depends on the rate of deleveraging at Firm 2 versus the rate that leverage increases at Firm 1 after the end of its deleveraging episode. In all cases, as long as some firms are increasing leverage while others are reducing deleverage, which is virtually always true in the real data, the event-time sample mean and median leverage ratios will yield downward-biased measures of the magnitude of deleveraging by the typical firm. The bias is large in our sample of deleveraging episodes following all-time ML peaks. For example, among the 4,476 firms with five or more years of post-peak data, it takes six years for the median firm to reach the post-peak trough and the median ML at that trough is 0.026. The event-time median ML ratios for these firms are 0.166 in the fifth year after peak and 0.157 in the sixth year. The latter figures give the incorrect impression that the typical firm deleverages to an ML ratio that is considerably above 0.000. As we document next, the bias is also large for deleveraging episodes that follow substantial leverage increases, that is, not limited to deleveraging from all-time peak. 6.3 Robustness analysis Table 13 replicates key aspects of our analysis on samples of firms with large leverage “spikes.” We define an ML spike to be an increase of at least 0.100 in ML in a given year, with no requirement that the ML resulting from the spike be the firm’s all-time peak. We analyze samples of the initial ML spike for each firm (first, second, fifth, and sixth columns) and of all ML spikes for all firms (all other columns). The first four columns examine cases with five or more years of post-spike data, while the last four examine cases with one to four years of post-spike data. Odd-numbered columns exclude cases where ML never decreases after the spike in question, that is, where no deleveraging occurs. Inclusion of such cases results in a small shift up in the typical deleveraging outcome, as can be seen by comparing each pair of samples labeled “exclude” and “include” in row 2 of the table. For brevity, we discuss only the results for initial spikes (first and fifth columns), but inspection of the table reveals that the other samples show qualitatively similar results. Table 13 Market leverage spikes and subsequent deleveraging 5 or more years of post-spike data 1 to 4 years of post-spike data Initial spike for each firm All spikes Initial spike for each firm All spikes Treatment of cases with no ML decline after spike: Exclude Include Exclude Include Exclude Include Exclude Include 1. Median ML in spike year 0.334 0.327 0.446 0.438 0.355 0.338 0.489 0.476 2. Median in year of trough after spike 0.063 0.084 0.089 0.107 0.165 0.229 0.241 0.317 3. Median ML in event time after spike $$\qquad$$ spike year $$+$$ 1 0.328 0.342 0.422 0.430 0.255 0.381 0.372 0.477 $$\qquad$$ spike year $$+$$ 2 0.302 0.325 0.384 0.401 0.246 0.367 0.332 0.434 $$\qquad$$ spike year $$+$$ 3 0.305 0.329 0.364 0.383 0.267 0.382 0.339 0.422 $$\qquad$$ spike year $$+$$ 4 0.303 0.333 0.354 0.376 0.276 0.356 0.315 0.385 $$\qquad$$ spike year $$+$$ 5 0.297 0.332 0.339 0.362 – – – – $$\qquad$$ spike year $$+$$ 6 0.285 0.312 0.327 0.348 – – – – $$\qquad$$ spike year $$+$$ 7 0.286 0.310 0.318 0.334 – – – – 4. Median Cash/TA in spike year 0.052 0.052 0.042 0.043 0.061 0.054 0.044 0.042 5. Median Cash/TA in trough year 0.101 0.093 0.090 0.086 0.078 0.062 0.066 0.054 6. Percentage of cases with $$\qquad$$ all debt paid off at trough 23.9% 21.6% 20.7% 19.2% 11.7% 7.0% 8.5% 5.5% $$\qquad$$ negative net debt at trough 52.7% 48.5% 47.4% 44.5% 35.6% 24.6% 27.5% 19.8% 7. Median percentage of deleveraging explained by $$\qquad$$ Debt repayment (DR) 66.5% 66.5% 57.3% 57.3% 51.6% 51.6% 42.3% 42.3% $$\qquad$$ DR and earnings retention (ER) 90.5% 90.5% 85.0% 85.0% 71.7% 71.7% 63.2% 63.2% $$\qquad$$ DR, ER, and share issuance (SI) 96.6% 96.6% 92.5% 92.5% 81.7% 81.7% 72.3% 72.3% 8. Median years from spike to trough 6 6 7 6 2 1 2 1 9. Number of firms 5,650 6,241 6,092 6,351 2,046 3,418 5,455 7,108 10. Number of ML spikes 5,650 6,241 15,862 17,117 2,046 3,418 6,578 10,262 5 or more years of post-spike data 1 to 4 years of post-spike data Initial spike for each firm All spikes Initial spike for each firm All spikes Treatment of cases with no ML decline after spike: Exclude Include Exclude Include Exclude Include Exclude Include 1. Median ML in spike year 0.334 0.327 0.446 0.438 0.355 0.338 0.489 0.476 2. Median in year of trough after spike 0.063 0.084 0.089 0.107 0.165 0.229 0.241 0.317 3. Median ML in event time after spike $$\qquad$$ spike year $$+$$ 1 0.328 0.342 0.422 0.430 0.255 0.381 0.372 0.477 $$\qquad$$ spike year $$+$$ 2 0.302 0.325 0.384 0.401 0.246 0.367 0.332 0.434 $$\qquad$$ spike year $$+$$ 3 0.305 0.329 0.364 0.383 0.267 0.382 0.339 0.422 $$\qquad$$ spike year $$+$$ 4 0.303 0.333 0.354 0.376 0.276 0.356 0.315 0.385 $$\qquad$$ spike year $$+$$ 5 0.297 0.332 0.339 0.362 – – – – $$\qquad$$ spike year $$+$$ 6 0.285 0.312 0.327 0.348 – – – – $$\qquad$$ spike year $$+$$ 7 0.286 0.310 0.318 0.334 – – – – 4. Median Cash/TA in spike year 0.052 0.052 0.042 0.043 0.061 0.054 0.044 0.042 5. Median Cash/TA in trough year 0.101 0.093 0.090 0.086 0.078 0.062 0.066 0.054 6. Percentage of cases with $$\qquad$$ all debt paid off at trough 23.9% 21.6% 20.7% 19.2% 11.7% 7.0% 8.5% 5.5% $$\qquad$$ negative net debt at trough 52.7% 48.5% 47.4% 44.5% 35.6% 24.6% 27.5% 19.8% 7. Median percentage of deleveraging explained by $$\qquad$$ Debt repayment (DR) 66.5% 66.5% 57.3% 57.3% 51.6% 51.6% 42.3% 42.3% $$\qquad$$ DR and earnings retention (ER) 90.5% 90.5% 85.0% 85.0% 71.7% 71.7% 63.2% 63.2% $$\qquad$$ DR, ER, and share issuance (SI) 96.6% 96.6% 92.5% 92.5% 81.7% 81.7% 72.3% 72.3% 8. Median years from spike to trough 6 6 7 6 2 1 2 1 9. Number of firms 5,650 6,241 6,092 6,351 2,046 3,418 5,455 7,108 10. Number of ML spikes 5,650 6,241 15,862 17,117 2,046 3,418 6,578 10,262 Market leverage (ML) is book debt divided by the sum of book debt and the market value of equity. A leverage spike is defined as an increase in a given year of 0.100 or more in ML. Rows 1 and 2, respectively, report the median actual ML ratio inclusive of the 0.100 leverage spike and the median actual ML ratio at the post-spike trough. The first, third, fifth, and seventh columns exclude observations for which ML never subsequently goes below the spike level. The second, fourth, sixth, and eighth columns include those observations. In those columns, the medians in rows 1 to 5 and 8 are based on samples in which those observations are included, but the medians in row 7 exclude them (because there is no deleveraging to be explained). Row 3 presents the median leverage ratios for the sample aligned in event time, with spike year$$+$$1 denoting the year after the spike, spike year$$+$$2 denoting the second year after the spike, etc. Table 13 Market leverage spikes and subsequent deleveraging 5 or more years of post-spike data 1 to 4 years of post-spike data Initial spike for each firm All spikes Initial spike for each firm All spikes Treatment of cases with no ML decline after spike: Exclude Include Exclude Include Exclude Include Exclude Include 1. Median ML in spike year 0.334 0.327 0.446 0.438 0.355 0.338 0.489 0.476 2. Median in year of trough after spike 0.063 0.084 0.089 0.107 0.165 0.229 0.241 0.317 3. Median ML in event time after spike $$\qquad$$ spike year $$+$$ 1 0.328 0.342 0.422 0.430 0.255 0.381 0.372 0.477 $$\qquad$$ spike year $$+$$ 2 0.302 0.325 0.384 0.401 0.246 0.367 0.332 0.434 $$\qquad$$ spike year $$+$$ 3 0.305 0.329 0.364 0.383 0.267 0.382 0.339 0.422 $$\qquad$$ spike year $$+$$ 4 0.303 0.333 0.354 0.376 0.276 0.356 0.315 0.385 $$\qquad$$ spike year $$+$$ 5 0.297 0.332 0.339 0.362 – – – – $$\qquad$$ spike year $$+$$ 6 0.285 0.312 0.327 0.348 – – – – $$\qquad$$ spike year $$+$$ 7 0.286 0.310 0.318 0.334 – – – – 4. Median Cash/TA in spike year 0.052 0.052 0.042 0.043 0.061 0.054 0.044 0.042 5. Median Cash/TA in trough year 0.101 0.093 0.090 0.086 0.078 0.062 0.066 0.054 6. Percentage of cases with $$\qquad$$ all debt paid off at trough 23.9% 21.6% 20.7% 19.2% 11.7% 7.0% 8.5% 5.5% $$\qquad$$ negative net debt at trough 52.7% 48.5% 47.4% 44.5% 35.6% 24.6% 27.5% 19.8% 7. Median percentage of deleveraging explained by $$\qquad$$ Debt repayment (DR) 66.5% 66.5% 57.3% 57.3% 51.6% 51.6% 42.3% 42.3% $$\qquad$$ DR and earnings retention (ER) 90.5% 90.5% 85.0% 85.0% 71.7% 71.7% 63.2% 63.2% $$\qquad$$ DR, ER, and share issuance (SI) 96.6% 96.6% 92.5% 92.5% 81.7% 81.7% 72.3% 72.3% 8. Median years from spike to trough 6 6 7 6 2 1 2 1 9. Number of firms 5,650 6,241 6,092 6,351 2,046 3,418 5,455 7,108 10. Number of ML spikes 5,650 6,241 15,862 17,117 2,046 3,418 6,578 10,262 5 or more years of post-spike data 1 to 4 years of post-spike data Initial spike for each firm All spikes Initial spike for each firm All spikes Treatment of cases with no ML decline after spike: Exclude Include Exclude Include Exclude Include Exclude Include 1. Median ML in spike year 0.334 0.327 0.446 0.438 0.355 0.338 0.489 0.476 2. Median in year of trough after spike 0.063 0.084 0.089 0.107 0.165 0.229 0.241 0.317 3. Median ML in event time after spike $$\qquad$$ spike year $$+$$ 1 0.328 0.342 0.422 0.430 0.255 0.381 0.372 0.477 $$\qquad$$ spike year $$+$$ 2 0.302 0.325 0.384 0.401 0.246 0.367 0.332 0.434 $$\qquad$$ spike year $$+$$ 3 0.305 0.329 0.364 0.383 0.267 0.382 0.339 0.422 $$\qquad$$ spike year $$+$$ 4 0.303 0.333 0.354 0.376 0.276 0.356 0.315 0.385 $$\qquad$$ spike year $$+$$ 5 0.297 0.332 0.339 0.362 – – – – $$\qquad$$ spike year $$+$$ 6 0.285 0.312 0.327 0.348 – – – – $$\qquad$$ spike year $$+$$ 7 0.286 0.310 0.318 0.334 – – – – 4. Median Cash/TA in spike year 0.052 0.052 0.042 0.043 0.061 0.054 0.044 0.042 5. Median Cash/TA in trough year 0.101 0.093 0.090 0.086 0.078 0.062 0.066 0.054 6. Percentage of cases with $$\qquad$$ all debt paid off at trough 23.9% 21.6% 20.7% 19.2% 11.7% 7.0% 8.5% 5.5% $$\qquad$$ negative net debt at trough 52.7% 48.5% 47.4% 44.5% 35.6% 24.6% 27.5% 19.8% 7. Median percentage of deleveraging explained by $$\qquad$$ Debt repayment (DR) 66.5% 66.5% 57.3% 57.3% 51.6% 51.6% 42.3% 42.3% $$\qquad$$ DR and earnings retention (ER) 90.5% 90.5% 85.0% 85.0% 71.7% 71.7% 63.2% 63.2% $$\qquad$$ DR, ER, and share issuance (SI) 96.6% 96.6% 92.5% 92.5% 81.7% 81.7% 72.3% 72.3% 8. Median years from spike to trough 6 6 7 6 2 1 2 1 9. Number of firms 5,650 6,241 6,092 6,351 2,046 3,418 5,455 7,108 10. Number of ML spikes 5,650 6,241 15,862 17,117 2,046 3,418 6,578 10,262 Market leverage (ML) is book debt divided by the sum of book debt and the market value of equity. A leverage spike is defined as an increase in a given year of 0.100 or more in ML. Rows 1 and 2, respectively, report the median actual ML ratio inclusive of the 0.100 leverage spike and the median actual ML ratio at the post-spike trough. The first, third, fifth, and seventh columns exclude observations for which ML never subsequently goes below the spike level. The second, fourth, sixth, and eighth columns include those observations. In those columns, the medians in rows 1 to 5 and 8 are based on samples in which those observations are included, but the medians in row 7 exclude them (because there is no deleveraging to be explained). Row 3 presents the median leverage ratios for the sample aligned in event time, with spike year$$+$$1 denoting the year after the spike, spike year$$+$$2 denoting the second year after the spike, etc. Table 3 documents that firms tend to deleverage from all-time peak to conservative leverage, while Table 13 indicates that conservative leverage is also the typical deleveraging outcome following ML spikes. Among cases with five or more years of post-spike data, median ML is 0.334 in the year of the spike (row 1) and it is 0.063 at the post-spike trough (row 2). Importantly, the post-spike event-time sequence of median ML ratios fails to capture the substantial size of the typical deleveraging (compare rows 2 and 3, Table 13). (See Section 6.2 for the explanation.) Moreover, consistent with Table 3’s analysis of deleveraging from peak, Cash/TA ratios typically double as ML declines from spike to subsequent trough (rows 4 and 5), with roughly one fourth of firms paying off all debt and about half attaining negative net debt (row 6). Debt repayment and earnings retention account for 90.5% of the reduction in ML for the median firm, while these two decisions plus share issuance account for 96.6% (row 7). Deleveraging from spike to trough takes six years for the median firm (row 8), which is the length of the typical deleveraging from peak by firms with at least five years of post-peak data (row 2, Table 10). Consistent with the peak-focused analysis in earlier sections of this paper, the extent of deleveraging is attenuated following ML spikes for which there are just a few years of post-spike data available. The median spike-inclusive ML is 0.355 (row 1, fifth column, Table 13), while median ML at the subsequent trough is 0.165 (row 2). Debt repayment, earnings retention, and share issuance continue to play an important role, accounting for 81.7% of the deleveraging by the median sample firm (row 7). Our findings are quite similar when we repeat the Table 13 analysis using book leverage (BL) rather than ML (see Table IA3 in the Internet Appendix). For example, in the BL analog to the analysis in the first column of Table 13, the median spike-inclusive BL ratio is 0.331, while median BL at the post-spike trough is 0.055. Median Cash/TA increases from 0.063 to 0.114, and debt repayment, earnings retention, and share issuance together account for 100.0% of the post-spike decline in BL for the median firm. In sum, the same qualitative picture emerges when we examine ML and BL following leverage spikes as we found in the analysis of deleveraging after peak ML. 7. Implications of Key Findings for Capital Structure Theories The firm-level deleveraging episodes we study are much larger and more focused on reaching low leverage than one would expect from the prior empirical literature. Most firms face financial trouble when they are at their all-time peak market leverage (ML), with almost 22% delisted due to distress within four years of peak. Distressed delists typically exit the sample with ML ratios well above the post-peak troughs of firms that are not delisted. Most of the latter firms deleverage to safe financial condition, with near-zero ML, high cash balances, and negative net debt. Debt repayment typically plays the main direct role in deleveraging. Earnings retention also plays an important direct role by raising the denominator of the ML ratio, especially at firms that initially have high ML ratios and/or that increase debt while reducing leverage. Share issuance typically has only a small direct impact on the ML ratio but, as with internal equity generated by retention, external equity’s direct impact is nontrivial at the minority of sample firms that increase debt while reducing ML. Among the many firms that repay debt, the dollar amounts of retention and share-issuance proceeds are typically large relative to the amount of debt repaid (and to post-deleveraging cash balances). Hence new equity capital constitutes an economically material indirect (funding-related) contribution to deleveraging by firms that repay debt. These findings indicate that credible theories of financial policy must be able to explain pervasive (large-scale) proactive deleveraging from peak to near-zero ML accompanied by decisions to bolster cash holdings materially. Such theories must also be able to explain why, among the many firms that pay off all debt, most drive their net-debt ratios deeply negative. Theories in which capital structure decisions are driven by financial-flexibility considerations can explain our main findings. We conclude the paper with a detailed discussion of this specific point and, more generally, of what our findings indicate about specific properties theorists will need to include when seeking to formulate models that can explain observed corporate financial policies. 7.1 Implications for traditional leverage targeting The widespread proactive deleveraging from peak to near-zero ML that we find is hard to reconcile with the idea that most firms have materially positive leverage targets of the type in traditional trade-off theories of capital structure. In traditional theories, all deviations from leverage targets are exogenous to managers, and all leverage changes undertaken by managers move leverage toward target. If traditional targeting behavior is at work and if leverage targets are materially positive, why do so many firms proactively overshoot those targets by large amounts as they reduce ML to zero or near-zero levels, while also driving net debt negative? With trivial transaction costs of distributing cash, firms could avoid overshooting of targets by buying back shares or paying a special dividend instead of accumulating large cash balances, as they actually do while deleveraging. Such transitory equity payouts could have kept ML ratios higher (and net debt positive) at most sample firms. Instead, firms chose to grow cash balances, pushing ML closer to zero and net debt negative. Thus, revealed preference indicates that most managers’ behavior is not driven by the objective of keeping leverage ratios close to materially positive targets. This point is reinforced by the fact that most firms also have near-zero ML at the trough before the peak. Moreover, although concerns about financial distress plausibly motivate firms to reduce ML from high levels, such concerns cannot explain proactive deleveraging to near-zero ML coupled with increases in cash holdings that push many firms deep into the negative-net-debt zone. 7.2 Leverage targeting when financial flexibility is valuable Pervasive deleveraging to near-zero ML is consistent with the very different (from traditional theories) target-related behavior in flexibility-based theories of capital structure. Because flexibility-based theories posit that debt has a marginal-issuance-cost advantage over equity and that debt capacity is finite, a firm’s capital structure target is the leverage ratio that optimally positions it to issue debt to meet future funding needs.9 If there are no tax, agency, or other benefits to having some debt permanently in the capital structure, the target is zero leverage. The target exceeds zero to the extent there are benefits to permanent debt. In either case, a firm that is currently at its leverage target will sometimes issue debt and voluntarily move away from that target. That is the nature of financial flexibility: it is valuable only when situations might arise in which firms choose to use it. Because debt capacity is a scarce resource, firms have incentives to deleverage back toward target after debt capacity is used. Such deleveraging restores the firm’s “dry powder,” that is, the option to issue debt again to meet new funding needs. The firm will eventually deleverage all the way to its target capital structure if (and only if) future funding needs and available resources permit. Even if a firm deleverages all the way to target, it will generally not choose to stay there. The reason is that new funding needs will tend to arise and lead to another transitory debt issuance, that is, a decision to move away from target coupled with incentives to follow that debt issuance by deleveraging to restore the option to borrow. Intuitively, the leverage targeting here conforms to that of a simple “credit-card” theory of debt. To see why, assume that firms do not have any motive to have debt as a permanent component of capital structure. In this case, the target balance on the credit card is zero because that provides maximal future funding capacity. Firms of course sometimes use their credit cards to meet funding needs, and so they will sometimes have large account balances and a currently limited ability to borrow more. However, as future earnings arrive, firms proactively deleverage (repay their debt) to restore the capacity to borrow. The point here is not that the deleveraging episodes we study exclusively reflect the restoration of financial flexibility, as other considerations could also be at work. For example, a decision to move from high ML to a conservative capital structure could plausibly reflect changes over time in managerial views of acceptable levels of corporate risk. After all, most firms are financially troubled at peak leverage and that experience may have made managers more conservative and thus encouraged proactive movement toward low leverage and high cash balances. On the other hand, many firms are not distressed at peak and they, too, tend to rebuild flexibility by reducing leverage and increasing cash balances substantially. It is difficult to explain the latter firms’ deleveraging simply by increased managerial fear attributable to a recent experience with financial trouble. In any case, the point here is that the leverage dynamics we observe are closer to the type of target-related behavior in the flexibility-based credit-card story than in traditional trade-off models. For example, we find that many sample firms proactively deleverage from all-time peak ML to negative net debt. Traditional tax versus distress cost models hold investment policy fixed and therefore cannot explain this regularity. With investment policy fixed, there is a cost (avoidable corporate taxes) and no benefit from ever having more cash than debt. 7.3 Cash balances and leverage Our evidence can thus also be read as indicating that traditional trade-off models of capital structure need to move beyond financial distress costs as a motive to limit leverage and also include a non-distress-related motive for cash-balance accumulation. The most obvious such motive is to acquire flexibility, with firms accumulating cash to meet possible future funding needs. However, agency problems or behavioral motives (e.g., managerial fear) could also lead to the choice of financial policies with ample cash holdings coupled with low leverage. 7.4 Earnings retention and internally generated equity The empirical corporate finance literature generally treats capital structure as the most important aspect of financial policy, with payout policy a largely separable issue of secondary importance. This empirical view clashes at the most basic level with the theoretical analyses of Stiglitz (1973), Myers and Majluf (1984), and others who emphasize the benefits of internally generated equity. In the latter studies, internal financing yields benefits because corporate retention of resources implies lower personal taxes (in an overall present-value sense) and mitigates asymmetric information problems associated with external financing. The two views clash because managers generate internal equity for a firm’s capital structure through the choice of payout policy and, in particular, through decisions to retain rather than pay out earnings. We interpret our findings on deleveraging and earnings retention as supporting capital structure theories in which there are advantages to internally generated equity. Decisions to retain large amounts of earnings, which are commonplace after firms reach ML peaks, are an empirically important means of effecting large-scale reductions in ML, especially from high ML ratios and for firms that increase debt while deleveraging. 7.5 Firms trade off restoring debt capacity (and growing cash balances) to bolster payouts Many sample firms increase dividends while deleveraging. Managers of these firms could have reduced ML more rapidly and, in some cases, to a lower level without upsetting shareholders with a dividend cut. All managers had to do was hold dividends constant and use the incremental retention to pay down debt and/or build up cash balances. Their decisions reveal that they willingly accept muted deleveraging in order to deliver larger payouts to shareholders. In other words, firms often treat payout considerations as important in their own right, and not as dominated by the benefits of rebuilding flexibility by deleveraging and accumulating larger cash balances. This emphasis on payouts may still reflect financial-flexibility considerations since, as DeAngelo and DeAngelo (2007) note, a demonstrated commitment to making equity payouts should help provide a firm with reliable future access to equity capital. 7.6 Elements of trade-off and pecking-order theories that matter Fama and French (2005) conclude that traditional trade-off and pecking-order theories both fail as stand-alone models of capital structure, and that researchers should focus on elements of each that help explain financing decisions. We find that deleveraging from financially troubled to safe condition pervades our sample. This finding suggests that the distress-cost side of the leverage trade-off in traditional theories is likely to be an important element of credible theories of financial policy. Our findings also suggest a significant role for financial flexibility, as the pecking-order theory was first to highlight in the literature. Theories in which flexibility is valuable plausibly explain why many firms do not simply reduce leverage to low, but positive, levels to avoid distress, and instead increase cash balances and thereby deleverage deep into the negative-net-debt zone. In the end, these two considerations plausibly merge into one, as both distress costs and a positive value to financial flexibility ultimately rest on firms having imperfect access to funding. 7.7 Permanent versus transitory debt Our data further suggest a need to take seriously the possibility that many firms view debt largely as a funding tool and do not see material (tax or agency) benefits from having debt permanently in the capital structure. The reason is that proactive deleveraging to near-zero ML and negative net debt is pervasive, and such deleveraging is incompatible with firms treating debt as a permanent source of capital. At a minimum, there is a need to recognize that most firms use debt as transitory capital, with transitory connoting “not-permanent” rather than “short-term.” 7.8 Key features of credible theories Viewed most broadly, our findings highlight the need for theories that (1) can explain why most firms proactively deleverage from peak to near-zero ML (and negative net debt) after having had similarly conservative financial policies before peak; (2) treat financial flexibility and the option to borrow as valuable, with debt used for transitory financing; (3) recognize the benefits of internally generated equity obtained through the retention of earnings; and (4) have important financial flexibility-related dependencies among leverage, cash-balance, and retention versus payout decisions. 7.9 Bottom line Our findings support capital structure theories in which financial flexibility is valuable, with firms deleveraging to restore the option to borrow, and they cast doubt on the idea that most firms have materially positive leverage targets. For helpful comments, we thank an anonymous referee, David Denis (editor), Gerard Hoberg, Arthur Korteweg, Rodney Ramcharan, Richard Roll, Michael Schwert, Berk Sensoy, Clifford Smith, Michael Weisbach, and Ivo Welch and brownbag seminar participants at Ohio State. An earlier version of this paper circulated under the title “Corporate Deleveraging.” Supplementary data can be found on The Review of Financial Studies web site. Appendix A.1 Contribution of Managerial Actions to Observed Deleveraging This appendix describes how we gauge the contribution of three decisions—repay debt, retain earnings, and/or issue shares—to the actual deleveraging by each sample firm. These calculations underlie the three categories of results reported in rows 9 and 10 of Table 3. For a given firm, ML(peak) and ML(trough) denote the actual market-leverage ratios at the beginning and end of the deleveraging episode. MVE(peak) and MVE(trough) denote the actual total market values of equity at the ML peak and trough, and Debt(peak) and Debt(trough) denote the actual values of debt that, respectively, are embedded in ML(peak) and ML(trough). We use the symbol HML(trough) to denote what the firm’s hypothetical market-leverage ratio would be at the trough if the only things that changed during the time between peak and trough are the specific managerial decisions in question. A.2 Hypothetical Deleveraging Due Solely to Debt Repayment For the debt repayment entry in row 9 of Table 3, we calculate HML(trough) under the assumption that the only thing that changes from peak ML is that the firm pays down its debt from Debt(peak) to Debt(trough). In this case, we first define HD $$=$$ Min[Debt(trough), Debt(peak)]. We then substitute HD into HML(trough) $$=$$ HD/[HD $$+$$ MVE(peak)]. Note that, if a given firm actually increases the dollar value of debt during its deleveraging, then HD $$=$$ Debt(peak). In this case, HML(trough) $$=$$ Debt(peak)/[Debt(peak) $$+$$ MVE(peak)] $$=$$ ML(peak) so that the hypothetical deleveraging magnitude due to debt repayment per se is nil. A.3 Hypothetical Deleveraging Due to Debt Repayment and Earnings Retention We first calculate HMVE(trough), the hypothetical market value of equity at the post-peak trough under the assumption that the only thing that has changed is the accumulation of additional earnings within the firm. Specifically, we set HMVE(trough) $$=$$ MVE(peak) plus the increase in retained earnings over the period from peak to trough. We then use the value of HMVE(trough) as an input to calculate HML(trough) $$=$$ Min[Debt(trough)/$$\{$$Debt(trough) $$+$$ HMVE(trough)$$\}$$, ML(peak)]. This calculation assumes a zero impact on the total market value of equity from (1) new share issuance, (2) equity payouts that are not netted out from retained earnings, (3) any element of normal stock price appreciation beyond that captured by retained earnings, (4) unanticipated changes in other managerial decisions, and (5) exogenous shocks to a firm’s opportunity set or to general capital market conditions. There are three other things to note about this calculation. First, it likely understates the true impact of retained earnings per se on the market value of equity because it makes no adjustment for compounded returns on any earnings retained in the years prior to attainment of the post-peak trough. On the other hand, it assumes that (uncompounded) retention adds dollar-for-dollar to the market value of equity, with no discount for any tax that will apply to future payouts that result from the hypothetical retention. Second, if the change in retained earnings is negative, we make no adjustment and instead treat the hypothetical equity value at the trough as equal to the value that prevailed at the peak, that is, HMVE(trough) $$=$$ MVE(peak). In this case, any leverage reduction from ML(peak) is due solely to the repayment of debt (as in the calculation for the debt repayment entry in row 9 of Table 3). Third, this calculation allows increases in the dollar value of debt over the deleveraging episode; that is, Debt(trough) $$>$$ Debt (peak) is admissible. In such cases, HML(trough) falls below ML(peak) only if the impact of retained earnings on the denominator of the leverage ratio outweighs the impact on the numerator of the debt increase. If the reverse holds, then the formula sets HML(trough) $$=$$ ML(peak) so that the hypothetical deleveraging magnitude is nil. A.4 Hypothetical Deleveraging Due to Debt Repayment, Earnings Retention, and Net Share Issuance In this case, we first take the retained earnings-inclusive value of HMVE(trough) described immediately above and add the change (from peak to trough) due to issuance of shares during the deleveraging episode. To obtain the latter estimate for each given firm, we proceed as follows. For each month in the interval from peak to trough, we take the change in the firm’s outstanding shares and value that change at the average of the beginning-of-month and end-of-month share prices (with all share-quantity changes and share prices treated on a consistent split-adjusted basis). We then sum these firm-specific monthly value increments over all months beginning with the month after peak and ending with the trough. The resultant sum is our estimate of the change in value of the equity base due to managerial decisions regarding net share issuance. We use the term net share issuance because this calculation will shrink the equity base when the firm repurchases stock and reduces the outstanding number of shares. (This approach assumes that repurchases have negligible impact on reported retained earnings and instead largely affect paid-in capital on the balance sheet.) We take the resultant value of HMVE(trough) and use it to generate HML(trough) following the formula given in the calculation for debt repayment and earnings retention. A.5 Percentage of Actual Deleveraging Attributable to Managerial Decisions Table 3 (rows 9 and 10) and various subsequent tables report the percentage of a firm’s actual deleveraging that is explained by managerial decisions to repay debt, issue shares, and/or retain earnings. To generate the relevant percentage figures for each firm, we use the inputs defined above in the following calculations. The actual deleveraging amount is [ML(peak) – ML(trough)]. The hypothetical deleveraging amount is [ML(peak) – HML(trough)]. If HML(trough) $$\leqslant$$ ML(trough), then the decisions account for 100.0% of the actual deleveraging. If HML(trough) $$=$$ ML(peak), then the decisions account for 0.0% of the actual deleveraging. For intermediate cases, the portion of deleveraging explained by the decisions being analyzed is the percentage equivalent of the ratio [ML(peak) – HML(trough)] $$\div$$ [ML(peak) – ML(trough)], which in these cases will fall strictly between 0.0% and 100.0%. Footnotes 1 Many prior studies ignore earnings retention and assume that security issuance/retirement decisions are the sole variables managers can use to alter leverage. A prominent example is Welch’s (2004) influential study, which decomposes time-series variation in ML ratios into endogenous and exogenous components. Because Welch does not consider the incremental amount of internally generated equity capital obtained through earnings retention, his decomposition understates the extent to which the evolution of ML is endogenous to managers. 2 For evidence of weak-to-moderate year-to-year mean reversion in leverage, see Fama and French (2002), Welch’s (2004), Leary and Roberts (2005), Flannery and Rangan (2006), Kayhan and Titman (2007), Huang and Ritter (2009), and Hovakimian and Li (2011). Welch’s (2004) concludes that firms do not issue or repurchase securities to counteract the mechanistic influence of stock price changes on market-leverage ratios. Hovakimian and Li (2011) critique the literature and conclude that the best available estimates imply weak mean reversion in leverage. See also Frank and Goyal (2008), Parsons and Titman (2008), and Graham and Leary (2011). 3 In Table 3, the incremental impact of retained earnings (93.7% minus 71.3% $$=$$ 22.4%) differs from the stand-alone impact of RE, which is 27.1% (not tabulated). Similarly, the incremental impact of share issuance in Table 3 (96.5% minus 93.7% $$=$$ 1.8%) is not equal to the stand-alone impact of SI, which is 6.9% (not tabulated). The difference between the incremental and stand-alone figures reflects that there are interaction effects among the three components of deleveraging. This makes intuitive sense because earnings retention and stock issuances provide funds that can be used for debt repayment. We discuss the latter issue in Section 3.6. For the current discussion, the key point is that the incremental contribution measures of RE and SI in Table 3 offer the same qualitative picture as the stand-alone measures: for the typical sample firm, debt repayment is considerably more important than retained earnings and stock issuance as a direct contributor to deleveraging. 4 Note that “sticky dividends” is not a conceptual property of payout/retention policy. Rather, it is an empirical description of dividend decisions by many, but not all, U.S. firms in Lintner’s (1956) day. It is of doubtful current relevance for many U.S. firms given the massive growth in share repurchases (Skinner 2008). It is also of doubtful relevance for firms based in some countries other than the United States, for example, Japan (Dewenter and Warther 1998). 5 Share-issuance proceeds and earnings retention contribute to funding in two subtly distinct ways. Issuance proceeds provide greater resources inside the firm that can be spent. Earnings retention is reduced through greater cash payouts, which imply fewer resources inside the firm that can be spent. The latter property holds even when, as is true of the Compustat data we use, measured retention reflects accounting earnings that differ from cash flow. 6 Denis and McKeon (2012, pp. 1902–3) specify three requirements for an observed increase in ML to be included in their sample of proactive leverage increases: (1) the current annual change in ML must be at least 0.100, (2) the post-increase ML ratio must be at least 0.100 above the firm’s target leverage ratio (estimated using industry leverage and other leverage determinants traditionally posited in the literature), and (3) the total dollar change in debt must be at least 90% of a scale-adjusted change in the firm’s total outstanding debt. 7 The variables in rows 8 and 9 are ML at the peak interacted with financial distress at peak: Row 8 has ML at peak conditional on the firm having an Altman z-score at peak $$<$$ 1.81, and row 9 has ML at peak conditional on the firm’s z-score at peak $$=$$ 1.81. Rows 10 to 12, 17, and 18 contain indicator variables that take the value one if a particular specified condition is met, and zero otherwise. The row 10 variable equals one if the firm’s z-score at peak ML $$<$$ 1.81. The row 11 variable equals one if the firm’s ML increase in the peak year satisfies the Denis and McKeon (2012) condition for a proactive increase in leverage. The row 12 variable equals one if the year of peak ML comes within six months of an NBER-specified recession. 8 In our sample, most firms increase ML by nontrivial amounts within a reasonably short time after reaching their post-peak ML troughs. This fact is consistent with the post-deleveraging use of financial flexibility by most sample firms. Among firms with five years of post-trough data, 35.9% have ML remain for at least five years in a bandwidth of 0.050 after reaching the ML trough. A large majority (84.8%) of these stable ML cases occur when the ML trough is 0.050 or lower. (The latter firms have ample unused debt capacity for at least five years, indicating they have not yet seen a need to tap their “dry powder” or to lever up to a materially positive target.) Moreover, it is rare to find ML rebounding upward from the ML trough and then stabilizing. We find 487 cases in which ML increases by at least 0.050 in the year after the trough (and in which the firm has an additional five years of data on Compustat). Only 2% of these 487 cases have ML stabilize in a bandwidth of 0.050 through the fifth year after the post-trough jump in ML. The analogous percentage is 3% for ML jumps (of size 0.050 or larger) that occur in the second or third years after the trough. The fact that a “rebound-and-stabilize” pattern almost never occurs in our data runs counter to the view that the ML trough is a minimum value that simply involves temporarily overshooting (on the downside) a higher target ratio. 9 In traditional trade-off theories, there is no option value for unused debt capacity. In flexibility-based theories, the key conditions that make unused debt capacity valuable and that lead to an option-inclusive leverage target are (1) a multiperiod setting with endogenous investment/operating policy, (2) debt capacity is a scarce resource, (3) debt is a relatively low cost means of raising outside funds, and (4) there is a (tax, liquidity, or agency) cost that deters unlimited stockpiling of cash balances (DeAngelo, DeAngelo, and Whited 2011, pp. 235–36). References Asquith, P., Gertner, R. and Scharfstein. D. 1994 , Anatomy of financial distress: An examination of junk-bond issuers. Quarterly Journal of Economics 109 : 625 – 58 . Google Scholar CrossRef Search ADS Brennan, M. 1971 . A note on dividend irrelevance and the Gordon valuation model. Journal of Finance 26 : 1115 – 21 . Danis, A., Rettl, D. and Whited. T. 2014 . Refinancing, profitability, and capital structure. Journal of Financial Economics 114 : 424 – 43 . Google Scholar CrossRef Search ADS DeAngelo, H., and DeAngelo. L. 2007 . Capital structure, payout policy, and financial flexibility. Working Paper , USC Marshall School of Business . https://papers.ssrn.com/sol3/papers.cfm?abstract_id=916093. Google Scholar CrossRef Search ADS DeAngelo, H., DeAngelo, L. and Whited. T. 2011 . Capital structure dynamics and transitory debt. Journal of Financial Economics 99 : 235 – 61 . Google Scholar CrossRef Search ADS DeAngelo, H., and Roll. R. 2015 . How stable are corporate capital structures? Journal of Finance 70 : 373 – 418 . Google Scholar CrossRef Search ADS Denis, D. 2011 . Financial flexibility and corporate liquidity. Journal of Corporate Finance 17 : 667 – 74 . Google Scholar CrossRef Search ADS Denis, D., and McKeon. S. 2012 . Debt financing and financial flexibility: Evidence from proactive leverage increases. Review of Financial Studies 25 : 1897 – 929 . Google Scholar CrossRef Search ADS Dewenter, K., and Warther. V. 1998 . Dividends, asymmetric information, and agency conflicts: Evidence from a comparison of the dividend policies of Japanese and U.S. firms. Journal of Finance 53 : 879 – 904 . Google Scholar CrossRef Search ADS Donaldson, G. 1961 . Corporate debt capacity . Washington, DC : Beard Books . Fama, E., and French. K. 2002 . Testing tradeoff and pecking order predictions about dividends and debt. Review of Financial Studies 15 : 1 – 33 . Google Scholar CrossRef Search ADS Fama, E., and French. K. 2005 . Financing decisions: Who issues stock? Journal of Financial Economics 76 : 549 – 82 . Google Scholar CrossRef Search ADS Fischer, E., Heinkel, R. and Zechner. J. 1989 . Dynamic capital structure choice: Theory and tests. Journal of Finance 44 : 19 – 40 . Google Scholar CrossRef Search ADS Flannery, M., and Rangan. K. 2006 . Partial adjustment toward target capital structure. Journal of Financial Economics 79 : 469 – 506 . Google Scholar CrossRef Search ADS Frank, M., and Goyal. V. 2008 . Tradeoff and pecking order theories of debt. In Handbook of corporate finance , volume 2 . Ed. Eckbo, B. 135 – 202 . Amsterdam : Elsevier . Frank, M., and Goyal. V. 2009 . Capital structure: Which factors are reliably important? Financial Management 38 : 1 – 37 . Google Scholar CrossRef Search ADS Gilson, S. 1997 . Transactions costs and capital structure choice: Evidence from financially distressed firms. Journal of Finance 52 : 161 – 96 . Google Scholar CrossRef Search ADS Gilson, S., John, K. and Lang. L. 1990 . Troubled debt restructurings: An empirical study of private reorganization of firms in default. Journal of Financial Economics 27 : 315 – 53 . Google Scholar CrossRef Search ADS Graham, J., and Harvey. C. 2001 . The theory and practice of corporate finance: Evidence from the field. Journal of Financial Economics 60 : 187 – 243 . Google Scholar CrossRef Search ADS Graham, J., and Leary. M. 2011 . A review of empirical capital structure research and directions for the future. Annual Review of Financial Economics 3 : 309 – 45 . Google Scholar CrossRef Search ADS Harford, J., Klasa, S. and Walcott. N. 2009 . Do firms have leverage targets? Evidence from acquisitions. Journal of Financial Economics 93 : 1 – 14 . Google Scholar CrossRef Search ADS Hovakimian, S., and Li. G. 2011 . In search of conclusive evidence: How to test for adjustment to target capital structure. Journal of Corporate Finance 17 : 33 – 44 . Google Scholar CrossRef Search ADS Huang, R., and Ritter. J. 2009 . Testing theories of capital structure and estimating the speed of adjustment. Journal of Financial and Quantitative Analysis 44 : 237 – 71 . Google Scholar CrossRef Search ADS Kayhan, A., and Titman. S. 2007 . Firms’ histories and their capital structures. Journal of Financial Economics 83 : 1 – 32 . Google Scholar CrossRef Search ADS Lang, L., Poulsen, A. and Stulz. R. 1995 . Asset sales, firm performance, and the agency costs of managerial discretion. Journal of Financial Economics 37 : 3 – 37 . Google Scholar CrossRef Search ADS Leary, M., and Roberts. M. 2005 . Do firms rebalance their capital structures? Journal of Finance 60 : 2575 – 619 . Google Scholar CrossRef Search ADS Lemmon, M., Roberts, M. and Zender. J. 2008 . Back to the beginning: persistence and the cross-section of capital structure. Journal of Finance 63 : 1575 – 608 . Google Scholar CrossRef Search ADS Lintner, J. 1956 . Distribution of incomes of corporations among dividends retained earnings, and taxes. American Economic Review 46 : 97 – 113 . Modigliani, F., and Miller. M. 1963 . Corporate income taxes and the cost of capital: A correction. American Economic Review 53 : 433 – 43 . Myers, S., and Majluf. N. 1984 . Corporate financing and investment decisions when firms have information that investors do not have. Journal of Financial Economics 13 : 187 – 221 . Google Scholar CrossRef Search ADS Parsons, C., and Titman. S. 2008 . Empirical capital structure: A review. Foundations and Trends in Finance 3 : 1 – 93 . Google Scholar CrossRef Search ADS Rajan, R., and Zingales. L. 1995 . What do we know about capital structure? Some evidence from international data. Journal of Finance 50 : 1421 – 60 . Google Scholar CrossRef Search ADS Skinner, D. 2008 . The evolving relation between earnings, dividends, and stock repurchases. Journal of Financial Economics 87 : 582 – 609 . Google Scholar CrossRef Search ADS Stiglitz, J. 1973 . Taxation, corporate financial policy, and the cost of capital. Journal of Public Economics 2 : 1 – 34 . Google Scholar CrossRef Search ADS Welch, I. 2004 . Capital structure and stock returns. Journal of Political Economy 112 : 106 – 31 . Google Scholar CrossRef Search ADS © The Author 2017. Published by Oxford University Press on behalf of The Society for Financial Studies. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)
http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png
The Review of Financial Studies
Oxford University Press
http://www.deepdyve.com/lp/oxford-university-press/corporate-deleveraging-and-financial-flexibility-jRhg0hGtUW