Dynamic Corporate Finance under Costly Equity Issuance

Patrick Bolton and Neng Wang

Two fundamental concepts in cor- Although tax distortions and bank- (3) equity issuance costs and cash carry porate finance are the net present value ruptcy costs are obviously relevant, they costs.3 A first, key result of our analy- (NPV) rule and the Modigliani-Miller cannot alone account for most observed sis is that investment is no longer deter- (MM) irrelevance proposition. When corporate financial decisions. They can- mined by equating the marginal cost of financial markets operate without fric- not explain why companies hold so much investing with the marginal addition to tions, when investors can trade securi- cash, their leverage dynamics, nor their the firm’s valuation from such capital, as ties that correspond to all relevant risks, payout, equity issuance, and investment in the neoclassical theory of investment. when investors and managers share the policies. We show in our research that the Instead, investment is determined by the same information, when incentives are cost of issuing equity is a key and practi- ratio of the marginal increase in the firm’s aligned, and when there are no tax distor- cally relevant distortion. Because of asym- value to the marginal value of cash. The tions, then corporate finance boils down metric information or incentive misalign- marginal cost of investing equals the mar- to a valuation exercise and a simple invest- ment, firms must incur costs when raising ginal product of capital, also known as ment decision rule: undertake all invest- external funds1 and these costs are higher marginal q, divided by the marginal value ments with a positive NPV. How com- for equity than for debt financing.2 of cash. panies and investments are financed is When firms face external financing When firms are flush with cash, the irrelevant. costs, they seek to avoid such financ- marginal value of cash is about one, so This characterization of financial ing. This is a key reason that firms retain that this equation is approximately the markets is frequently taken as approxi- earnings and accumulate cash (corpo- same as the equation under MM irrel- mately valid; a plausible and convenient rate savings). With Hui Chen, we ana- evance. But when firms are close to run- simplification even if it poorly reflects lyze a dynamic model with three main ning out of internal funds, or close to the reality. Corporate income taxation, the building blocks: (1) an investment rule limit of their credit line, the marginal interest tax shield for debt, and bank- based on the marginal value of incremen- value of cash is much larger than one, so ruptcy costs are often the only deviations tal capital investment relative to its cost, that marginal product may need to yield a from this view that are considered when (2) cash, equity, and a credit line as fund- much higher return, and optimal invest- explaining corporate financing choices. ing sources (together with hedging), and ment may be far lower, than the level pre-

Patrick Bolton is the Barbara and David Zalaznick Professor of Business at Columbia University and a visiting professor at Imperial College London. An NBER research associate in the Corporate Finance Program, he is a past president of the American Finance Association, a fellow of the Econometric Society and the American Academy of Arts and Sciences, and a corre- sponding fellow of the British Academy. He has coauthored Contract Theory (2005) with Mathias Dewatripont and The Green Swan: Central Banking and Financial Stability in the Age of Climate Change (2020) with Morgan Despres, Luiz Pereira Da Silva, Frédéric Samama, and Romain Svartzman.

Neng Wang is the Chong Khoon Lin Professor of Real Estate and Finance at Columbia Business School. He is also a research associate in the NBER’s Asset Pricing Program and a senior research fellow at the Asian Bureau of Finance and Economic Research. Wang is an associate editor of The Journal of Finance and was an editor in finance atManagement Science. His research interests include corporate finance, asset pricing, and macroeconomics. Wang was born and raised in China. He received his BS in chemistry from Nanjing University, an MS in chemistry from the California Institute of Technology, an MA in international relations from the University of California, San Diego, and his PhD in finance from Stanford University in 2002. He lives in Scarsdale, NY, with his wife and three children.

10 NBER Reporter • No. 2, June 2021 dicted under MM neutrality. age cash-to-assets ratio for US public cor- or boom times, when external financing Figure 1 illustrates the sizable value porations more than doubled from 1980 costs are affordable (cheap), firms opti- destruction that a financial crisis can to 2006, and remained elevated after the mally time their equity offerings and issue cause, as firms are shut out of capital mar- 2008 financial crisis.6 equity even when there is no immediate kets and can only rely on internal funds to need for external funds. Along with the continue their operations.4 Panels A and Market Timing and timing of equity issuance by firms with B show that firm value is increasing and Financial Crises low cash holdings in good market con- concave in cash holdings, that the mar- ditions, there is also optimal timing of ginal value of cash always exceeds one, Our model predicts that cash hold- stock repurchases by firms with large cash and that it is very large when the firm ings increase when earnings volatility holdings. Just as firms with low cash hold- runs out of cash. Panel C shows that firms rises, but this is not an adequate explana- ings seek to take advantage of low costs substantially cut investment and engage tion for the rise in corporate savings. A of external financing to raise more funds, in very costly fire sales firms with high hold- when liquidity is low. ings will be inclined The firm values a dollar Corporate Cash Balances, Investment, and Valuation to disburse their cash in hand at about $30 through stock repur- A. Firm value/capital ratio B. Marginal value of cash and sells about 60 per- 1.6 30 chases when financing cent of its productive First-best 25 conditions improve. 1.4 capital at a significant 20 This result is consistent value discount when it 1.2 15 with the finding that is close to being inef- aggregate equity issu- Liquidation 10 ficiently liquidated, 1.0 ances and stock repur- 5 in sharp contrast to chases are positively 0.8 0 8 the predictions of the 0.00 0.05 0.10 0.15 0.20 0.25 0.00 0.05 0.10 0.15 0.20 0.25 correlated. When the neoclassical theory of Cash/captial ratio Cash/captial ratio perceived probability investment. Finally, C. Investment-capital ratio D. Investment-cash sensitivity of a crisis rises, firms Panel D reveals how 0.4 20 invest more conser- First-best nonlinear and non- 15 vatively, issue equity monotonic invest- 0.0 sooner, and delay pay- 10 ment-cash sensitivity Liquidation outs to shareholders, all -0.4 can be, indicating that 5 to increase cash hoards investment-cash sensi- that will help them tivity is a poor measure -0.8 0 through the impend- of how financially con- 0.00 0.05 0.10 0.15 0.20 0.25 0.00 0.05 0.10 0.15 0.20 0.25 ing crisis. Finally, we strained a firm is.5 Cash/captial ratio Cash/captial ratio demonstrate that firms’ A second key Source: Bolton P, Chen H, and Wang N. NBER Working Paper 14845, and published as “A Unified Theory of Tobin's q, risk premia have two result concerns the Corporate Investment, Financing, and Risk Management” Journal of Finance, 66(5), October 2011, pp 1545–78 components: pro- firm’s optimal cash- ductivity and financ- inventory policy. That Figure 1 ing. Both risk pre- involves continuous management of cash more plausible explanation is the risk of mia change substantially with firms’ cash reserves through adjustments in invest- a financial crisis, which causes a jump in holdings, especially in a crisis when exter- ment, asset sales, and corporate hedging the cost of external financing and possibly nal financing conditions are poor. between two barriers: a lower bound at even a financial market shutdown. which the firm must tap external financ- With Chen, we further explore how Real Options and ing after exhausting all its cash reserves, firms’ financial policies are affected by Financial Flexibility and an upper bound at which the firm anticipation of random financial crises.7 has accumulated enough cash that it is We show that during such a crisis, firms Real-options theory, which applies safe to pay out any additional earnings. delay payout, cut investment, and engage when investments are lumpy and irrevers- Our model provides insights into how in fire sales of assets even when their ible, is an important subfield of corporate these bounds depend on factors such as productivity remains unaffected, all to finance that generally assumes that firms the growth rate and volatility of earn- avoid incurring prohibitive equity issu- operate in an MM environment. With ings, external financing costs, and capi- ance costs. This is especially true when Jinqiang Yang, we show that the presence tal adjustment costs. It can thus provide a firm enters the crisis with low cash of external financing costs fundamentally part of an explanation for why the aver- reserves. We also find that in normal alters the value and exercising decisions

NBER Reporter • No. 2, June 2021 11 associated with real options.9 To avoid performance and to the outside options its target leverage. These predictions are incurring external financing costs, firms of key employees. To retain risk-averse clearly counterfactual. delay investment until they have sufficient employees, the company optimally com- However, when we incorporate funds, and mostly finance their invest- pensates them by smoothing their con- equity issuance costs, the model yields ments with internally generated funds, sumption and limiting their risk exposure. plausible average leverage outcomes and consistent with the empirical evidence. We show that the objective of corpo- leverage dynamics. First, and somewhat In our model, investment, financ- rate risk management is not achieving an paradoxically, it is optimal for compa- ing, payout, and abandonment policies all optimal risk-return profile for investors; nies to target lower leverage when they depend on both earnings fundamentals they can do that on their own. Rather, risk face higher equity issuance costs. Indeed, and the firm’s cash holdings. We show that management is designed to offer optimal when it is costly to issue equity, it is best when cash holdings are depleted — fol- risk-return profiles to risk-averse, under- to avoid incurring such costs too often, lowing a crisis, for example — low invest- diversified, key employees. The com- which is achieved by keeping leverage ment persists even when earnings funda- pany is, in effect, both the employer and low to be able to cover a future loss mentals fully recover. After a crisis, firms the asset manager for its key employees. by borrowing, which is cheaper. Second, are in repair mode, seeking to rebuild Indeed, corporations invest 40 percent the company’s leverage increases fol- their internal funds. Also, firms favor of their liquid savings in risky financial lowing a loss and decreases following a investments with front-loaded earnings, assets, and less-constrained firms invest profit realization. Leverage can then only and payout policy is different depending more in the market portfolio.12 increase in response to earnings losses. on whether the firm is in a growth or a We further show that when compa- When the company attains its low lever- mature phase. In a mature phase a more nies are severely financially constrained age target any additional profit is paid profitable firm pays out more, while in a they cut compensation, reduce invest- out, and when leverage reaches the com- growth phase it pays out less. ment, engage in asset fire sales, and reduce pany’s debt capacity any additional loss hedging positions, with the primary either triggers a costly recapitalization via Managing Keyman Risk objective of surviving by honoring liabili- equity issuance or — when the jump loss ties and retaining key employees.13 is very large — a default. When leverage In addition to the cost of raising is close to the recapitalization target, the external funds, moral hazard is an impor- Leverage Dynamics under expected change in leverage is negative, so tant source of financial constraints. With Costly Equity Issuance that leverage tends to revert to the recapi- Yang, we explore a dynamic model where talization target. But when leverage passes the source of moral hazard is inalienabil- An important lesson from dynamic a certain threshold, the expected change ity of human capital10 — what is com- models of corporate finance is that “capi- in leverage is positive, so that the com- monly referred to as ”keyman risk” in tal structure is not static, but rather evolves pany enters a leverage death spiral. the tech industry to describe the risk that over time as an aggregation of sequential These leverage dynamics are consis- key employees could leave the firm.11 It decisions.”14 With Yang, we build on the tent with the empirical evidence pointing is often noted that tech companies stand work of Christopher Hennessy and Toni to the heterogeneity of corporate lever- out in terms of their cash holdings. We Whited and show how leverage dynam- age of firms with similar characteristics.16 explain these tech company cash policies ics can be naturally explained by com- Companies, in effect, behave like house- in terms of mitigation of keyman risk. panies’ efforts to avoid incurring equity holds with credit card debt, except that How do tech companies retain their issuance costs.15 We consider a company they also have an option to issue exter- most valuable engineers? Essentially by that can issue equity and short-term debt, nal equity to deleverage. As credit card offering enough deferred state-contingent facing both cash-flow diffusion and jump revolvers, firms pay down their debt when compensation. We show that the larger shocks. As in the MM trade-off theory they receive a positive earnings shock, the company’s cash holdings and bor- widely taught in MBA classes, when the and they increase their debt when they rowing capacity, the greater its ability to company faces no equity issuance costs it have no option to do otherwise, consis- retain talent by making credible com- always stays at its target leverage, defined tent with empirically observed leverage pensation promises. We also describe the as the point at which the benefits from dynamics.17 company’s optimal risk management pol- debt financing are equal to expected bank- icy, showing how the company’s idiosyn- ruptcy costs. In our model, debt has a net Dynamic Trade-offTheory cratic and aggregate risk exposures can funding advantage over equity because under Costly Equity Issuance be set to reduce both the cost of retain- shareholders are impatient. When mak- ing talent and the cost of financing. In ing a profit, the company uses it to pay With Chen, we add equity issuance our model, physical capital is illiquid and down debt to the extent that it stays at its costs to the standard dynamic trade-off depreciates randomly. The firm faces risk target leverage, and when making a loss it theory model of capital structure.18 An with respect both to its future financial raises just enough new equity to return to important additional cost of debt financ-

12 NBER Reporter • No. 2, June 2021 ing in this expanded model is debt service: Risk Management,” Bolton P, Chen H, March 2019, and The Journal of Finance debt payments drain the firm’s cash hold- Wang N. NBER Working Paper 14845, 74(3), 2019, pp. 1363–1429. ings, which increases the risk of incurring April 2009, and The Journal of Finance Return to Text equity issuance costs. Also, realized earn- 66(5), 2011, pp. 1545–1578. 12 “Precautionary Savings with Risky ings are separated in time from payouts Return to Text Assets: When Cash Is Not Cash,” to shareholders, so that savings have both 4 Ibid. Figure 2. Duchin R, Gilbert T, Harford J, a corporate tax, when savings are inside Return to Text Hrdlicka C. The Journal of Finance the firm, and a personal tax component, 5 “Do Investment-Cash Flow 72(2), 2017, pp. 793–852. when savings are outside the firm. In this Sensitivities Provide Useful Measures Return to Text setting, standard measures of the net tax of Financing Constraints?” Kaplan SN, 13 “Boarding a Sinking Ship? An benefits of debt are no longer valid. Zingales L. The Q uarterly Journal of Investigation of Job Applications to This framework can be extended Economics 112(1), 1997, pp. 169–215. Distressed Firms,” Brown J, Matsa DA. beyond the traditional corporate setting. “The Corporate Propensity to Save,” The Journal of Finance 71(2), 2016, pp. With Ye Li and Yang, we show that costly Riddick LA, Whited TM. The Journal 507–550. equity issuance also plays a critical role in of Finance 64(4), 2009, pp. 1729–1766. Return to Text understanding the dynamics of a bank’s Return to Text 14 “Presidential Address: Collateral and balance sheet, bank valuation, and the 6 “Why Do US Firms Hold So Much Commitment,” DeMarzo P. The Journal effects of equity capital and leverage reg- More Cash Than They Used To?” Bates of Finance 74(4), 2019, pp. 1587–1619. ulation.19 We develop a dynamic theory TW, Kahle KM, Stulz RM. The Journal Return to Text of banking in which the role of depos- of Finance 64(5), 2009, pp. 1985–2021, 15 “Leverage Dynamics and Financial its is akin to that of productive capi- and “Is There a US High Cash Holdings Flexibility,” Bolton P, Wang N, Yang tal in the neoclassical theory of invest- Puzzle after the Financial Crisis?” J. NBER Working Paper 26802, ment for nonfinancial firms. We show Pinkowitz LF, Stulz RM, Williamson February 2020. This research builds on that deposits create value for well-capital- RG. The Ohio State University Fisher “How Costly is External Financing? ized banks. However, the marginal value College of Business Working Paper No. Evidence from a Structural Estimation,” of deposits can turn negative for under- 2013-03-07. Hennessy CA, Whited TM. The capitalized banks, as further inflows of Return to Text Journal of Finance 62(4), 2007, pp. deposits may require the bank to raise 7 “Market Timing, Investment, and 1705–1745. more costly equity capital to comply with Risk Management,” Bolton P, Chen H, Return to Text leverage regulations. Our predictions on Wang N. NBER Working Paper 16808, 16 “Do Firms Rebalance Their Capital bank valuation and dynamic asset-liabil- February 2011, and Journal of Financial Structures?” Leary MT, Roberts MR. ity management are broadly consistent Economics 109(1), 2013, pp. 40–62. The Journal of Finance 60(6), 2005, with the evidence, and our model offers Return to Text pp. 2575–2619, and “Back to the new insights into the dynamics of bank- 8 “The Timing of Financing Decisions: Beginning: Persistence and the Cross‐ ing in a low interest rate environment. An Examination of the Correlation in Section of Corporate Capital Structure,” In sum, our research shows that Financing Waves,” Dittmar AK, Dittmar Lemmon ML, Roberts MR, Zender JF. avoiding future costly equity issuance is RF. Journal of Financial Economics 90(1), The Journal of Finance 63(4), 2008, pp. a key motive driving various aspects of 2008, pp. 59–83. 1575–1608. dynamic corporate financial behavior. Return to Text Return to Text 9 “Investment under Uncertainty with 17 “Corporate Deleveraging and Financial Constraints,” Bolton P, Wang Financial Flexibility,” DeAngelo H, 1 “Corporate Financing and N, Yang J. NBER Working Paper 20610, Gonçalves AS, Stulz RM. The Review Investment Decisions When Firms Have July 2019, and Journal of Economic of Financial Studies 31(8), 2018, pp. Information That Investors Do Not Theory 184, November 2019, 10.1016. 3122–3174. Have,” Myers SC, Majluf NS. Journal Return to Text Return to Text of Financial Economics 13(2), 1984, pp. 10 “A Theory of Debt Based on the 18 “Debt, Taxes, and Liquidity,” Bolton 187–221. Inalienability of Human Capital,” Hart P, Chen H, Wang N. NBER Working Return to Text O, Moore J. The Q uarterly Journal of Paper 20009, March 2014. 2 “The Capital Structure Puzzle,” Myers Economics 109(4), 1994, pp. 841–879. Return to Text SC. The Journal of Finance 39(3), 1984, Return to Text 19 “Dynamic Banking and the Value pp. 574–592. 11 “Optimal Contracting, Corporate of Deposits,” Bolton P, Li Y, Wang N, Return to Text Finance, and Valuation with Inalienable Yang J. NBER Working Paper 28298, 3 “A Unified Theory of Tobin’s q, Human Capital,” Bolton P, Wang N, December 2020. Corporate Investment, Financing, and Yang J. NBER Working Paper 20979, Return to Text

NBER Reporter • No. 2, June 2021 13