YIELD STRUCTURE OF TAXABLE VS. NONTAXABLE BONDS

• Published date: 01 June 1985
• Date: 01 June 1985
• Author: Geoffrey Booth,Moon Kim
• DOI: http://doi.org/10.1111/j.1475-6803.1985.tb00391.x

The Journal of Financial Research • Vol. VIII,

NO.2·

Summer 1985

YIELD STRUCTURE OF TAXABLE VS.

NONTAXABLE BONDS

Moon Kim *

and

Geoffrey Booth*

Abstract

In his 1976 presidential address to the American Finance Association, Miller shows

that

the equilibrium marginal personal tax rate on riskless bond income is equal to the marginal

corporate tax rate. In the presence of risk, he and, subsequently, others suggest

that

the

theoretical equilibrium occurs when the personal tax rate is less

than

the corporate tax rate.

This study investigates empirically these relationships by examining the yield ratio of non-

taxable to taxable debt at various risk levels. Both the riskless and risk propositions are con-

firmed.

I. Introduction

Under the assumptions of riskless bonds, no tax arbitrage,

and

the existence of

personal taxes on

bond

income

but

not on stock income, Miller (1977) shows

that

the

equilibrium marginal personal tax rate on bond income is equal to the marginal cor-

porate tax rate. With risky bonds, the equilibrium condition occurs when the personal

tax rate on bond income is below the corporate tax rate, aphenomenon

that

results

from loss of the tax shield of interest in the event of bankruptcy. In more rigorous

analyses, DeAngelo

and

Masulis (1980a), Taggart (1980),

and

Barnea, Haugen,

and

Senbet (1981) confirm the existence of this risky debt equilibrium condition.

Trzcinka (1982), after performing empirical tests on the Miller proposition of risk-

less bonds using a

random

intercept coefficient model, concludes

that

the yield ratio of

municipal to corporate bonds is, as Miller predicts, close to .52. He claims

that

the

random

coefficient model adequately controls for the relative risk associated with

bond ratings between taxable

and

nontaxable bonds. Nevertheless, the direct causes

for the randomness of

the

intercept

term

are left unexplained. On the other hand,

Litzenberger

and

Rolfo (1984) show

that

the implied marginal income tax rate esti-

mated

from U.S. bonds fluctuates widely over time.

The purpose of this study is to examine empirically the risky debt equilibrium hy-

pothesis

that

maintains

that

the yield ratio is related to default risk; the higher the

level of default risk in

the

bond

market, the higher the yield ratio of nontaxable to

taxable bonds. Also, the higher the possibility of bankruptcy, the smaller the differ-

ence between the yields of these two types of bonds. Confirmation of this hypothesis

supports the contention

that

the yield ratio of various nontaxable to taxable bonds

systematically changes from

the

riskless yield ratio (one minus the marginal tax rate)

to higher values in response to changes in the factor

that

is associated with the default

risk of bonds. Thus, this study provides some explanation on the relative yield struc-

ture between nontaxable

and

taxable bonds of different levels of riskiness.

*Syracuse University.

95