cash flows and leverage adjustments現(xiàn)金流和杠桿調(diào)整

1、Cash Flows and Leverage Adjustments,organizationAbstract and KeywordsSection 1 introduction.Section 2 presents some basic empirical models of corporate leverage, describes data sources, and explains how we compute tar

2、get leverage ratios for each firm-year. Section 3 shows some distinguishing features of the approach Include the importance of distinguishing between under- and over-levered firms’ leverage adjustments.Section 4 intro

3、duces the paper’s major innovation. We explain why operating cash flows affect a firm’s cost of making leverage adjustments, and modify a standard partial adjustment model to reflect the interaction between a firm’s cash

4、 flow needs and its capital structure adjustments. Section 5 analyzes the robustness of our resultsSection 6 extends the model in Section 4 to test whether financial constraints and market conditions affect adjustment

5、speeds. We find that adjustment speeds vary plausibly with both cross-sectional and inter-temporal variables, supporting our partial adjustment model of capital structure adjustment. The final section summarizes results

6、 and discusses their implications for capital structure theories.,AbstractRecent research: Emphasized the impact of transaction costs on firm leverage adjustments. This paper: 1. Cash flow realizations can provide op

7、portunities to adjust leverage at relatively low marginal cost. Cash flow features affect not only the leverage target, but also the speed of adjustment toward that target. 2.Adjustment speeds is driven by an economical

8、ly meaningful concept: adjustment costs. Accounting for this fact produces adjustment speeds that are significantly faster than previously estimated in the literature. 3.We also analyze how both financial constraints an

9、d market timing variables affect adjustments toward a leverage target.Keywords Leverage Target Speed of adjustment Adjustment costs,1.Introduction(10 paragraphs)1-5(research background)6-9(results of th

10、is paper)10 (organization of this paper.),Introduction(1-5),Do firms have leverage targets? Almost all research in this arena concludes that firms do have targets, but that the speed with which these targets are re

11、ached in unexpectedly slow.How quickly do they approach these targets? The speed with which these targets are reached is unexpectedly slow.What are the drivers and impediments of the targets? Adjustment cos

12、ts The firm’s incentive to access capital markets for other reasons. A firm’s cash flow realization can substantially affect the cost of making a leverage adjustment, regardless of whether the firm is raisi

13、ng or distributing external funds. Two stylized examples illustrate the joint effect of adjustment costs and cash flows on observed leverage adjustments.,Two stylized examples illustrate the joint effect of adjustment c

14、osts and cash flows on observed leverage adjustments.,,2.Basic leverage models and data,Basic leverage models：A standard partial adjustment model,We revise (1) to separate a firm’s leverage change into a passive, mechani

15、cal component and an active adjustment:,Eqs. (1) and (2) can then be estimated using OLS, with bootstrapped standard errors to account for the generated regressor (Pagan, 1984).OLS( with bootstrapped standard errors自舉標(biāo)準(zhǔn)

16、誤)Bootstrapping is a nonparametric approach for evaluating the distribution of a statistic based on random sample. （An alternative method for estimating the standard errors when the theoretical calculation is complicate

17、d or not available）,GMM( generalized method of moments廣義矩估計(jì))在隨機(jī)抽樣中，樣本統(tǒng)計(jì)量將依概率收斂于某個(gè)常數(shù)。這個(gè)常數(shù)又是分布中未知參數(shù)的一個(gè)函數(shù)。即在不知道分布的情況下，利用樣本矩構(gòu)造方程（包含總體的未知參數(shù)），利用這些方程求得總體的未知參數(shù)。GMM 不需要知道隨機(jī)誤差項(xiàng)的準(zhǔn)確分布信息，允許隨機(jī)誤差項(xiàng)存在異方差和序列相關(guān)，因而所得到的參數(shù)估計(jì)量比其他參數(shù)估計(jì)方法更有效。,Da

18、ta：Compustat(標(biāo)準(zhǔn)普爾數(shù)據(jù)庫（全球財(cái)務(wù)分析資料庫）)1965-2006CRSP(Center for Research in Security Prices)Outliers: winsorized(1-99%)Book leverage: decomposing the active and passive pieces is more straightforward.,Table 1,3.Initial e

19、stimation results( table 2),Industry capital expenditures is defined as the Fama-French industry-year average capital expenditures normalized by the previous period’s book assets.Fama-French:Fama和French 1992年對美國股票市場決定不

20、同股票回報(bào)率差異的因素的研究發(fā)現(xiàn)，股票的市場的beta值不能解釋不同股票回報(bào)率的差異，而上市公司的市值、賬面市值比、市盈率可以解釋股票回報(bào)率的差異。Fama and French 認(rèn)為，上述超額收益是對CAPM 中β未能反映的風(fēng)險(xiǎn)因素的補(bǔ)償。 The first to interact adjustment speed measures with cash flows,and thus evaluate the joint ef

21、fect of transaction costs and cash flow needs on firm’s adjustments towards target.,4.The effect of cash flow on capital market adjustment costs,,,Excessdev=(|Dev|-|CF|)*DevLarger(r1),,,Dev=L*-L,CF( cash flow),,,Overlap

22、=|CF|*DevLarger(r2),,,Dev=L*-L,CF,,,ExcessCF=(| CF |-|Dev|)*(1-DevLarger)(r4),Overlap=|Dev|*(1-DevLarger)(r3),Hypothesize that: r2≈r3>r1>r4,Table 3 baseline adjustment speeds with decomposition,One of the contr

23、ibutions,result: r2≈r3>r1>r4,Table 4 alternative definition of operating cash flows(Eq.4),5.Robustness,穩(wěn)健性檢驗(yàn)檢驗(yàn)的是實(shí)證結(jié)果是否隨著參數(shù)設(shè)定的改變而發(fā)生變化，如果改變參數(shù)設(shè)定以后，結(jié)果發(fā)現(xiàn)符號(hào)和顯著性發(fā)生了改變，說明不是robust的，需要尋找問題的所在。一般根據(jù)自己文章的具體情況選擇穩(wěn)健性檢驗(yàn)：1. 從數(shù)據(jù)出

24、發(fā)，根據(jù)不同的標(biāo)準(zhǔn)調(diào)整分類，檢驗(yàn)結(jié)果是否依然顯著；2. 從變量出發(fā)，從其他的變量替換，如：公司size可以用total assets衡量，也可以用total sales衡量；3. 從計(jì)量方法出發(fā)，可以用OLS, FIX EFFECT, GMM等來回歸，看結(jié)果是否依然robust；,Robustness,we re-estimate our target leverage ratios with year dummies to all

25、ow time series variation at the macroeconomic level to alter firms’ target leverage positions for the corresponding fiscal year. We make three further adjustments to our definition of the firm’s cash flow realization

26、(Eq. (4)) First, |CF|= |CF|-working capital Second, |CF|= |CF|-cash dividends paid in the previous year Third, |CF|= |CF|- debt maturing in the current fiscal year,Table 5 robustness,6.Financial constrain

27、ts and market timing,6.1Modifications to the partial adjustment models Financial constraintsMarket timing,6.2Financial constraints results,6.3.Financial timing results,Conclusion of section 6,Adjustment speeds va

28、ry plausibly with both cross-sectional and intertemporal variables supporting our partial adjustment model of capital structure adjustment.,7.Summary and conclusion,Overall, our empirical results are consistent with the

29、trade-off hypothesis of capital structure: firms have targets, and wish to return to those targets when costs make it optimal to do so. Partial adjustment models yield theoretically sensible results about how a firm’s ch

30、aracteristics and market conditions affect observed leverageThe benefits and costs of adjustment vary with the sign of the firm’s leverage gap (over-levered firms generally adjust more quickly), its operating cash flow,

31、 its investment opportunities, its access to capital markets, and some elements of market conditions. Further research should continue to investigate the underlying determinants of adjustment costs and benefits across d