Effects of call provisions and sale and leaseback restrictions on agency costs of debt.
The agency problem that exists between stockholders and bondholders of a corporation arises from the conflict of interests between these two major stakeholders of the firm. A reduction in firm value due to this conflict is an agency cost of debt. To the extent that stockholders have limited liability, they would have an incentive to change the structure and the composition of the firm's assets in a manner which would result in a gain at the expense of bondholders. According to the risk incentive agency problem of Barnea, Haugen, and Senbet (1980), subsequent to issuing bonds, stockholders and management have an incentive to take on lower-valued, riskier projects, resulting in a wealth transfer from debt holders to stockholders.
Stockholders have an incentive to substitute risky and lower-valued assets for safe and tangible assets to maximize their wealth even though debtholders' wealth and firm value would decline. Fang and Zhong (2004) and Larsen (2006) find strong evidence of firms engaging in risk-shifting behavior. The sale and lease-back arrangement is such a method of asset substitution. Stockholders and managers can sell a physical asset and invest the proceeds in a high-risk, high-return project. The asset sold remains available simply by leasing it back; thus, the continuous use of the asset for regular business operation is not interrupted. Nonetheless, bondholders face elevated risk, resulting in a decline in the value of their bonds. Being wary of this wealth transfer to stockholders, bondholders seek to include protective covenants in bond indentures to deter stockholders and managers from implementing such an asset substitution. To the extent that such protective covenants restrict asset substitution via sale and lease-back arrangements, the ability of stockholders and managers to increase the risk of the firm is deterred.
Stockholders and bondholders hold differing perspectives on the use of sale and leaseback arrangements. Stultz and Johnson (1985) demonstrate that issuing secured debt reduces underinvestment agency costs and increases the gains to stockholders as the market value of secured debt is higher than that of unsecured debt. Stockholders, therefore, would like to execute a sale and lease-back of some assets that have been financed by issuing unsecured debt because lease-back of an asset is similar to issuing secured debt in that legal claim to the asset is given to the lender (i.e. the lessor). In a nutshell, existing unsecured debt holders are faced with a new breed of secured debt holders (i.e. lessors) who have higher priority in getting paid than unsecured debt holders. Thus, in an effort to protect their interests, unsecured debt holders desire that explicit restrictions on sale and lease-back arrangements be placed in debt contracts.
As postulated by Taggart and Bodie (1978) and Barnea, Hauge, and Senbet (1980), risk incentive agency costs can be reduced by issuing callable bonds because the call option value (i.e., stockholders' wealth) declines as management shifts to riskier projects that reduce firm value. These theories, however, are not supported by empirical results from Crabbe and Helwege (1994), who conclude that call features do not reduce the agency costs of debt.
In this study, the authors test whether callable bonds have lower agency costs of debt than non-callable bonds. The paper also examines whether an explicit contractual restriction on risk incentive-triggered transactions reduces the agency costs of debt.
Taggart and Bodie (1978), Barnea, Haugen, & Senbet, (1980) and Thatcher (1985) examine how a call feature can reduce the agency costs of debt. When bondholders appear to reap most of profit from the low risk, albeit profitable, projects, stockholders would not be interested in investing in these types of projects, resulting in underinvestment. This particular agency cost can be reduced by callable bonds. By including early call provisions on the bonds, stockholders capture a larger portion of the gain from undertaking low-risk but profitable projects by limiting the bondholders' gain to just the call premium. Stockholders' incentive to increase their wealth at the expense of bondholders by investing in high risk, albeit unprofitable, projects can also be lessened by call features because the value of the stockholders' call option declines with a fall in firm value due to the high-risk albeit unprofitable projects.
An empirical paper by Crabbe and Helwege (1994) fails to support reduction in the agency cost of debt due to a call feature. First, they note that agency theories such as asymmetric information, asset substitution, and the underinvestment problem are difficult to distinguish empirically; then, they test the theories for agency cost reduction from the issuance of callable bonds. Overall, they find little support for any of the theories suggesting that call features can mitigate the agency problem.
Secured debt has an inherent contractual protection for creditors in that assets are pledged as collateral until the debt is fully redeemed. Stultz and Johnson (1985) demonstrate how secured debt moderates the contracting cost associated with mitigating the underinvestment agency problem. Secured debt holders' first claim to the profits of low-risk albeit profitable projects limit the gains realized by the outstanding unsecured debt holders. Stockholders end up with a wealth gain due to an increase in firm value resulting from a higher market price for the new secured debt. Shareholders' manipulation of debt structure induces another form of conflict of interest between secured bondholders and unsecured bondholders. By having them pit against each other, shareholders gain as firm value increases due to adding senior debt to the existing debt structure.
Smith and Warner (1979), argue that secured debt precludes asset substitution of risky assets for existing assets, thus decreasing the administrative and enforcement costs of debt contracts. Myers and Majluf (1984) argue that investor uncertainty about firm assets in place can be reduced by the use of secured debt. Therefore, agency costs of debt can be reduced because the tangible assets pledged for secured debt provide collateral. A financial lease contract is essentially secured debt in that the lender (i.e., lessor) has legal claim to the leased assets at the time of the contract. Thus, a financial lease can be used like a secured debt to reduce agency costs of debt as it precludes asset substitution and reduces asset uncertainty.
Bhojraj and Sengupta (2003) show that corporate governance mechanisms are linked to higher bond ratings and lower bond yields. Using probit regression, they find that governance mechanisms can reduce default risk by mitigating agency costs and monitoring managerial performance and by reducing information asymmetry between the firm and the lenders. Klock, Mansi, and Maxwell (2005) find that anti-takeover provisions reduce the cost of debt financing.
OBJECTIVE OF STUDY
The purpose of this paper is to empirically test for the effect of two contractual arrangements (i.e. call feature and restriction on the sale and lease-back arrangements) on the reduction of agency costs of debt. That is, the authors empirically test whether callable bonds have lower agency costs of debt than non-callable bonds. Also examined is whether an explicit restriction on risk incentive-based transactions effectively reduces the agency costs of debt.
Agency costs of debt can be reduced by a call feature because the call limits the benefit to bondholders from low-risk but positive NPV projects to just the call premium. The dependent variable of this study, the bond rating, which serves as a proxy for the agency costs of debt in an inverse manner, depends upon the following variables: amount raised, maturity, coupon rate, and call feature. Testing for statistical significance of the contracting indicator variable will reveal the effect of call feature on agency costs of debt. The linear regression model is
[Y.sub.i] = [alpha] + [P.sub.1]XH + [P.sub.2]X2i + P3X3i + P4X4i + Si (1)
where [Y.sub.i] = bond rating (an inverse proxy for agency costs of debt)
[X.sub.1i] = amount of issue
[X.sub.2i] = maturity
[X.sub.3i] = coupon rate
[X.sub.4i] = 1 for callable bonds and 0 for non-callable bonds (contracting variable).
DATA COLLECTION AND VARIABLES
All data were compiled directly from the Securities and Exchange Commissions' EDGAR database which lists the prospectuses for all recent and past long term debt issues from each firm studied. Bond indentures of the 204 sampled bond issues, registered with the SEC in 2006 and 2007 by the Dow Jones thirty blue chip industrial firms, were examined for call features and the trustee-stipulated explicit prohibition on the disposition of assets through the sale of assets and/or a sale and lease-back agreement. The sample has 150 callable bonds and 54 non-callable bonds. Other bond features such as amount raised, maturity, coupon rate, and bond rating were collected as well. Average ratings of callable bonds and non-callable bonds (not reported here) are 6.35 and 6.25 respectively; the difference is not statistically significant.
The contractual feature of whether or not the bond indenture has an explicit restriction is quantified by using an indicator variable: 1 for explicit restriction and 0 for implicit (i.e. no) restriction. Compared to 35.6 percent of sample bonds in the study of Smith and Warner (1979), 41.7 percent (85 out of 204) of bonds in our study were found to have an explicit constraint on the firm's disposition of assets through either the sale of assets or a sale and lease-back agreement.
Bond ratings are proxies for agency costs that are highly correlated with default risk (Crabbe & Helwege, 1994). Our basic conjecture is that bond ratings are inversely related to the agency costs of debt, which, in turn, are directly related to the probability of including restrictive covenants to protect bondholders' interest. The hypothesis is that the implicit sub-sample (i.e. bonds without explicit restriction) has low enough default risk for bondholders not to require explicit restriction with sufficient trust in issuers to behave prudently. All of the sample bonds were investment-grade at the time of issue. For bond ratings, the following conversion scale is used to assign a numerical value to each rating: AAA+ = 12, AAA = 11, AAA- = 10, AA+ = 9, AA = 8, AA- = 7, A+ = 6, A = 5, A- = 4, BBB+ = 3, BBB = 2, BBB- = 1.
The sampled bonds in our study have an average bond rating of 6.28, approximately equivalent to A+. Descriptive statistics for the two subsamples are given in Tables 2 and 3. The explicit group has an average of 6.06 and the implicit group has an average of 6.47. This comparison is consistent with the conjecture that the lower default risk of the implicit group contributes to higher bond ratings than that of explicit group of bonds.
The total amount of capital to be raised through a bond issue should be a significant contributor to the agency costs of debt. The bond issue will increase the total amount of debt, and with more debt in capital structure, Jensen and Meckling (1976) argue that shareholders have more incentive to alter investment projects so that they can expropriate bondholders' wealth. Explicit restrictions on asset disposition reduce the default risk of issuing firms. Thus, the hypothesis is that the larger the amount borrowed, the more likely an explicit restriction is to be included in the debt contract.
For the sampled bonds, the average amount of capital borrowed by issuing bonds is $679 million dollars ($303 million for implicit contracting bonds and $1,210 million for the explicit contracting group). As posited, on average, bond issues with explicit restriction are associated with a larger amount of capital raised than bond issues with an implicit only restriction.
Maturity should be a key variable in explaining the inclusion of explicit restrictive covenants in the bond indenture. The model of Barnea, Haugen, & Senbet, (1980) suggests that long-term securities provide less discipline on borrowing firms. Hence, long-term bonds should carry higher agency costs. It is predicted, therefore, that the probability of including explicit restriction on asset disposition increases as the maturity of bonds increases. The average maturity of the sampled bonds is 17.27 years. However, the direction of the difference in average maturities for the two subsamples is opposite to our prediction (14.72 years for the explicit group and 19.09 years for the implicit group).
Finally, the coupon rate should help explain the inclusion of restrictive covenants. A high coupon rate should be negatively related to the probability of explicit restriction due to the accelerated cash flow to bondholders. Then it is expected that the higher the coupon rate, the lower the probability of including explicit restrictions on disposition of assets. Consistent with this conjecture, the average coupon rate for the total sample is 5.85 percent, 5.71 percent for the explicit group and 5.94 percent for the implicit group.
Effect of Call Features on Agency Costs of Debt
The linear regression results reveal the agency costs of debt (served by bond rating as an inverse proxy) are related to each explanatory variable. The focus is on the call feature variable, [X.sub.4i]. A statistically significant coefficient for the contracting variable would indicate the variable has an effect on the agency costs of debt. The ordinary least square results are shown in Table 4.
As shown in Table 4, the dependent variable (bond rating, as an inverse proxy of agency costs of debt) is positively related to maturity, and is negatively related to coupon rate and call feature. With the exception of amount of issue, coefficients of all independent variables are statistically significant at 5 percent. Interpretation of this result is as follows: the maturity coefficient, with a p-value of almost zero, is negatively related to the agency costs of debt, such that the longer the maturity, the higher the bond ratings, i.e., the lower the agency costs of debt. This result contradicts the common notion that a long maturity is positively correlated with high default risk, which would be accompanied by a lower bond rating. The statistically significant positive coefficient indicates that long-term bonds are issued by firms that have low agency costs of debt. The coupon rate is found to be significant and negatively related to the bond rating, and, hence, positively related to the agency costs of debt. The coefficient for the coupon rate is negative and statistically significant with a p-value close to zero and indicates that the bonds with low coupon rates are rated more highly by the rating agencies. Firms with low agency costs of debt (i.e., high bond ratings) can raise capital in bond markets at low coupon rates. Most noteworthy, the binary call feature variable (i.e., 1 for callable bonds and 0 for non-callable bonds) is negatively related to the bond rating, and its coefficient is statistically significant with a p-value close to zero. This result indicates that a call feature reduces bond ratings by 1.30589 on the conversion scale. This negative effect of call feature on the agency costs of debt is further revealed by the 95 percent confidence interval for p4, the coefficient of call feature, X4i: -1.80699 [less than or equal to] [[beta].sub.4] [less than or equal to] -0.80479. Thus, with 95 percent confidence, callable bonds have ratings that are by somewhere between 1.807 and 0.805 lower, on the average, than non-callable bonds. The negative relationship between bond ratings and call features indicates that firms with low agency costs of debt have a penchant for issuing non-callable bonds. Most surprising is to find that callable bonds are rated lower (i.e., higher agency costs of debt) than non-callable bonds by about one to two notches on the conversion scale. The prediction of the various agency theories related to the positive role of a call feature in reducing the agency costs of debt is not supported by this study. In essence, the bond ratings, used as an inverse proxy for agency costs of debt, are negatively related to the call feature; thus, callable bonds are perceived to be riskier due to higher agency costs of debt.
Effect of Call Feature and Contracting Variable on Agency Costs of Debt
Stockholders and bondholders hold contrasting perspectives on the sale and lease-back arrangement: bondholders' main concern is whether the debt contract has well-stipulated covenants that will protect their wealth from the sale and lease-back arrangement. Conversely, shareholders view the arrangement as beneficial because the sale and lease-back arrangement is essentially equivalent to issuing secured debt that increases firm value and helps shareholders gain at the expense of unsecured debt holders.
Stockholders can avoid debt financing while expropriating bondholders' wealth through a sale and lease-back transaction. With this type of financial lease, a physical asset owned by a firm is sold, but leased back simultaneously. There is no interruption in the use of the asset for regular business operations. Ownership of the asset, however, is transferred to the lessor. Bondholders' wealth is jeopardized if stockholders invest the proceeds of the sale of the asset in higher risk, higher return projects. Stockholders would gain at the expense of bondholders. Stockholders will benefit if the new projects are successful. On the other hand, bondholders face a higher than originally expected risk of default. Being wary of this possible loss in bond values, bondholders often require that a protective covenant be included in the bond indenture explicitly restricting the sale and lease-back transaction on particular assets of the borrowing firm. The first role of a protective covenant is to prevent managers and shareholders from taking value-reducing actions that could be privately optimal because they expropriate bondholders (Tirole, 2006). In short, this view takes restriction of sale and lease-back as an arrangement that can reduce agency costs of debt.
Smith and Wakeman (1985) assert that lease contracts can be viewed as a strong form of secured debt where the lender receives legal claim to the secured assets at the time of the loan. A sale and lease-back arrangement involves a financial lease contract for an asset that is sold by stockholders but is leased back, achieving an uninterrupted production process. Unsecured debt holders are concerned about possible wealth expropriation by stockholders who can replace a firm-owned tangible asset with a leased asset. From the standpoint of bondholders, a superior quality asset owned by the firm is converted into an inferior leased asset, and unsecured debt holders are faced with increased risk. Through this arrangement of sale and lease-back, lessors, as secured debt holders, are given higher priority of redemption than that of unsecured debt holders. There would be a new conflict of interest between outstanding unsecured debt holders and newly joining secured debt holders (i.e., lessors involved in the sale and lease-back arrangements). According to this view, restriction of the sale and lease-back transaction would reduce agency costs due to conflicts of interest between outstanding unsecured debt holders and lessors, but would increase the agency costs of debt by preventing stockholder from implementing sale and lease-back arrangements that would result in gain through reduced contracting costs associated with issuing unsecured debt and the higher market value of lease that is essentially equivalent to secured debt.
To examine how restriction of sale and lease-back arrangement affect agency costs of debt, a new contracting indicator variable (1 for explicit restriction and 0 for implicit restriction on sale and lease-back) is added to our regression model. Testing for statistical significance of the contracting indicator variable will reveal the effect of contractual restriction on agency costs of debt. The linear regression model is [Y.sub.i] = [alpha] + [[beta].sub.1] x [X.sub.1i] + [[beta].sub.2] x [X.sub.2i] + [[beta].sub.3] x [X.sub.3i] + [[beta].sub.4] x [X.sub.4i] + [[beta].sub.5] [X.sub.5i] + [[epsilon].sub.i], where [Y.sub.i] = bond rating (inverse proxy for agency costs of debt), [X.sub.1i] = amount of issue, [X.sub.2i] = maturity, [X.sub.3i] = coupon rate, X4i = call feature binary variable and contracting variable [X.sub.5i] = 1 for explicit restriction and 0 for implicit restriction on sale and lease-back arrangement.
The results of regression analysis show how the agency costs of debt (as proxied by bond ratings) are related to each explanatory variable. The focus is on whether adding the contracting variable changes the relation between call features and the agency costs of debt, and how the contracting indicator variable, [X.sub.5i] is related to agency costs of debt. The results of regression analysis are shown in Table 5.
As shown in Table 5, the relationships between the dependent variable (bond ratings, as an inverse proxy of agency costs of debt) and the independent variables are the same as in Table 4 in terms of the sign and statistical significance: bond rating is positively related to maturity, and is negatively related to coupon rate and call feature variable. Interestingly, the call feature variable is still negatively related to bond rating with a p-value close to zero despite the addition of the binary contracting variable that is negatively related to the agency costs of debt, and its coefficient is statistically significant with p-value 0.0125. This result indicates that the explicit restriction reduces bond ratings by 0.6027. This negative effect of explicit restriction on the agency costs of debt can be expressed in the 95 percent confidence interval for p5, the coefficient of contracting variable, [X.sub.5i]: -1.07402 [less than or equal to] [[beta].sub.5] [less than or equal to] -0.13132. Thus, with 95 percent confidence, the conclusion is that bonds under explicit restriction tend to have ratings that are between 0.13 and 1.07 lower, on average, than those only under implicit restriction.
With an explicit restriction on the sale and lease-back arrangement, there is no possibility of an increase in agency costs arising from a conflict of interest between the existing unsecured debt holders and new lessors; however, reduction of agency costs of debt via contracting cost reduction and higher market value of secured debt is not feasible due to the restriction being placed in the bond indenture. Thus, agency costs of debt increase (i.e., the bond rating falls) as the sale and lease-back arrangement is explicitly restricted in bond indenture and firms with low agency costs of debt are positioned to raise debt capital relatively free of explicit restriction placed in bond indenture. Bonds issued under implicit restriction appear to perform better than bonds under explicit restriction in mitigating the conflict of interest between bondholder and shareholder as indicated by a bond rating differential between 0.13 and 1.07. Basically, the negative relationship between bond rating and explicit restriction implies that the firms issuing debt under explicit restrictions would continue to suffer from high agency costs of debt associated with contracting and enforcement costs of restrictive covenants and the low market value of unsecured debt; whereas, bonds issued under implicit restriction are perceived to be less risky in terms of expropriation of bondholders' wealth through a sale and lease-back transaction.
Rational debt contracting results in protective covenants intended to protect bondholders from possible wealth expropriation by managers and shareholders. This paper explores the relationship between the agency costs of debt and bond features, with emphasis on call features and contracting scheme (either explicit or implicit restriction on sale and lease-back arrangements).
Bond rating (used as an inverse proxy for the agency cost of debt) is found to be negatively related to coupon rate in both tests. This result is interpreted in the following way: the firms with low agency costs of debt (i.e., high bond ratings) can raise capital in the bond market at low coupon rates.
The results also indicate that the longer the maturity, the higher the bond rating is, i.e., the lower agency costs of debt. This result contradicts the common belief that a long maturity is positively correlated with high default risk that would be revealed by lower bond ratings. The statistically significant positive coefficient indicates that long-term bonds are issued by firms that have low agency costs of debt.
The call feature variable is negatively related to the bond rating. Hence, the call feature is positively related to the agency costs of debt. The results indicate that call features reduce bond ratings by 1.30589. The negative relationship between bond ratings and call features implies that firms with low agency costs of debt are positioned to raise debt capital by issuing non-callable bonds. It is significant to learn that callable bonds are rated lower (i.e., higher agency costs of debt) than non-callable bonds. Our results run counter to the prediction of risk incentive that postulates the positive role of a call feature in reducing agency costs of debt.
The binary contracting variable (i.e., 1 for explicit restriction and 0 for implicit restriction) is also negatively related to the bond rating, and the negative relationship indicates that bonds issued with such a restrictive covenant on the sale and lease-back arrangement tend to be issued by firms that suffer from high agency costs of debt. That is, bonds issued under implicit (i.e., no) restrictions are perceived to be less risky due to lower agency costs of debt. The results are consistent with the notion that without an explicit restriction on asset substitution via sale and lease-back, the firm gains due to saving contracting/enforcement costs and a higher market value of the lease obligation as secured debt. The implication is that implicit restrictions reduce agency costs more efficiently than do explicit restrictions. Under this interpretation, the mere issuance of bonds without explicit restrictions in the debt contract will result in the firm being viewed as behaving prudently by following its own self-imposed restriction and, hence, higher bond ratings. This interpretation contradicts the common notion that the agency costs of debt would be reduced by the inclusion of explicit restrictions on the disposition of assets in the debt contract.
Barnea, A., Haugen, R.A., & Senbet, L.W. (1980). A rationale for debt maturity structure and call provisions in the agency theoretic framework. Journal of Finance, 35, 1223-1234.
Bhojraj, S., & Sengupta, P. (2003). Effect of corporate governance on bond ratings and yields: the role of institutional investors and outside directors. Journal of Business, 76, 455-476.
Crabbe, L., & Helwege, J. (1994). Alternative tests of agency theories of callable bonds. Financial Management, 23, 3-20.
Fang, M., & Zhong, R. (2004). Default risk, firm's characteristics, and risk shifting. (Working paper 04-21). New Haven, Connecticut: Yale School of Management.
Jensen, M.C., & Meckling, W.H. (1976). Theory of the firm: managerial behavior, agency costs, and capital structure. Journal of Financial Economics, 3, 305-360.
Klock, M.S., Mansi, S. A., & Maxwell, W. F. (2005). Does Corporate Governance Matter to Bondholders? Journal of Financial and Quantitative Analysis, 40, 693-719
Larsen, P.T. (2006). Default risk, debt maturity and levered equity's risk shifting incentives. (Working paper). Aarhus, Denmark: University of Aarhus, School of Economics and Management.
Myers, S., & Majluf, N. (1985). Corporate financing and investment decisions when firms have information that investors do not have, Journal of Financial Economics, 13, 187-221.
Smith, C., & Wakeman, L. (1985). Determinants of corporate leasing policy. Journal of Finance, 40, 895-908.
Smith, C., & Warner, J. (1979). On financial contracting: an analysis of bond covenants. Journal of Financial Economics, 7, 117-161.
Stultz, R., & Johnson, H. (1985). An analysis of secured debt, Journal of Financial Economics, 14, 501-521.
Taggart, R., & Bodie, Z. (1978). Future investment opportunities and the value of the call provision on a bond. Journal of Finance, 33, 1187-1200.
Thatcher, J.S. (1985). The choice of call provision terms: evidence of the existence of agency costs of debt. Journal of Finance, 40, 549-561.
Tirole, J. (2006). The theory of corporate finance. Princeton, NJ: Princeton University Press.
R. Barry Ellis
Maryellen P. Epplin
University of Central Oklahoma
About the Authors:
Minje Jung is a Professor of Finance at the University of Central Oklahoma. This is part of a series of research papers on the agency costs of debt and contracting theory. The first paper is forthcoming in the fall 2009 issue of the Journal of Business and Behavioral Science.
R. Barry Ellis is an Assistant Professor of Finance at the University of Central Oklahoma. His research interest is in corporate governance.
Jeffrey Kautz received his B.B.A. in Finance from the University of Central Oklahoma in May 2009.
Maryellen P. Epplin is a Professor of Finance at the University of Central Oklahoma. Her current research interests include issues in corporate finance and investments, and the use of technology in financial analysis.
Table 1 Descriptive Statistics for Variables (All Bonds) Statistic Amount ($) Maturity (years) Coupon (%) Rating Mean 679,200,344 17.27 5.8454 6.28 Median 500,000,000 12.5 5.975 6 Mode 750,000,000 30 6.00 6 Range 4,747,253,000 42 3.90 10 Minimum 2,747,000 3 4.25 1 Maximum 4,750,000,000 45 8.15 11 # of bonds 204 204 204 204 Table 2 Descriptive Statistics for Variables (Implicit Group) Statistic Amount ($) Maturity (years) Coupon (%) Rating Mean 3.03E+08 19.09 5.944 6.44 Median 27,639,000 17 6.00 6 Mode 7.5E+08 30 6.00 6 Range 3E+09 27 3.90 3 Minimum 2,747,000 3 4.25 5 Maximum 3E+09 30 8.15 8 # of bonds 119 119 119 119 Table 3 Descriptive Statistics for Variables (Explicit Group) Statistic Amount ($) Maturity (years) Coupon (%) Rating Mean 1.21E+09 14.72 5.71 6.06 Median 9E+08 10 5.60 6 Mode 7.5E+08 10 5.50 5 Range 4.73E+09 41 3.35 10 Minimum 25,000,000 4 4.35 1 Maximum 4.75E+09 45 7.70 11 # of bonds 85 85 85 85 Table 4 Regression of Bond Ratings on Bond Feature Variables and Call Feature SUMMARY OUTPUT Regression Statistics Multiple R 0.389303 R Square 0.151557 Adjusted R Square 0.134503 Standard Error 1.446965 Observations 204 ANOVA df SS MS Regression 4 74.42559819 18.6064 Residual 199 416.6479312 2.093708 Total 203 491.0735294 F Significance F Regression 8.886816 1.27128E-06 Residual Total Coefficient Standard Error t -Stat Intercept 11.01257 1.302849672 8.452676 Amount -6.9E-11 1.25364E-10 -0.55334 Maturity 0.055456 0.013811541 4.015169 Coupon -80.123 23.91703556 -3.35004 Call -1.30589 0.254114437 -5.13898 P-value Lower 95% Upper 95% Intercept 6.01E-15 8.443403532 13.5817293 Amount 0.580648 -3.1658E-10 1.7784E-10 Maturity 8.41E-05 0.028219919 0.08269144 Coupon 0.000967 -127.286382 -32.959675 Call 6.57E-07 -1.80699043 -0.8047852 Table 5 Regression of Bond Ratings on Bond Feature Variables, Call Feature Variable and Contracting Variable SUMMARY OUTPUT Regression Statistics Multiple R 0.4218433 R Square 0.1779518 AdjustedR Square 0.157193 Standard Error 1.4278721 Observations 204 ANOVA df SS MS Regression 5 87.38742 17.4774 Residual 198 403.6861 2.038819 Total 203 491.0735 F Significance F Regression 8.572357 2.28E-07 Residual Total Coefficient Standard t-Stat Error Intercept 11.46237 1.29798 8.83096 Amount 1.026E-10 1.4E-10 0.72632 Maturity 0.054326 0.01364 3.98385 Coupon -85.58816 23.7008 -3.6112 Call -1.27397 0.25108 -5.07395 Contracting -0.60267 0.23902 -2.5214 P-value Lower 95% Upper 95% Intercept 0 8.902736 14.022 Amount 0.468503 -1.76E-10 3.812E-10 Maturity 0 0.027435 0.0812181 Coupon 0.000386 -132.3265 -38.84985 Call 0 -1.769105 -0.778834 Contracting 0.012476 -1.074016 -0.131316
|Printer friendly Cite/link Email Feedback|
|Author:||Jung, Minje; Ellis, R. Barry; Kautz, Jeffrey; Epplin, Maryellen P.|
|Publication:||International Journal of Business, Accounting and Finance (IJBAF)|
|Date:||Jun 22, 2010|
|Next Article:||Significance of beta and financial measures in predicting the riskiness of S&P 500 stocks during the downturn of year 2008.|