Www.icra.in Files MoneyFinance JanMarch2003corp-Bondmarket

8/10/2019 Www.icra.in Files MoneyFinance JanMarch2003corp-Bondmarket

1/27

25

Money

Finance

A Study of the Indian CorporateBond Market

SUCHISMITA BOSE AND

DIPANKOR COONDOO

*

AbstractWithin any countrys capital ma rket, it is essential tha t t here

exist a w ell-developed bond ma rket w ith a sizeab le corpora te bond

segment alongside the ba nking system, so that the market mechanism

ensures that funds flow in accordance w ith the productivity of indi-

vidual investments a nd the ma rket exerts a competitive pressure on

commercial b anks lending t o priva te business and helps improve the

efficiency o f t he entire capital ma rket. Further, the debt ma rket must

emerge as a stab le source of finance to business when the eq uity

markets are volatile. However, most countries do not have corporate

bond markets compara ble in efficiency w ith their equity ma rkets, as the

secondary market for corpora te debt is mostly O ver-the-C ount er (and/or

telephonic), ra ther tha n exchange tra ded, a nd it is extensively domi-

nated by a few institutional investors and professional money ma nag-

ers. The mar ket for non-sovereign debt (part icularly, the corpo rat e debt

segment) in India a lso has a number of shortcomings: a prima ry ma rket

structure w here private placements, sans ma nda tory credit ra tings,

dominate in an overwhelming manner, lack of transparent market

making, a nd a tendency on the part of institutiona l investors to hold

securities to ma turity. The seconda ry market is thus prone to suffer

from low liquidity a nd fra gmentat ion and the conseq uent pricing

anomalies.

In this paper, w e make an a ttempt to understa nd the nature and

extent of imperfection of the Indian ma rket for corpora te bonds using

available data on seconda ry ma rket t rading channelled t hrough the

major sto ck exchanges. We exa mine some a spects of the ma rket w hich

include depth of the market in terms of f requency of tra ding of o ut-

standing bonds; composition of the market in terms of tra ding of debt

of va rious risk categories as indicat ed by t heir credit ra tings; relat ion-

ship betw een Yield-to-Ma turity (YTM ) and volat ility o f return; nat ure

Suchismita Bose is with the Monetary Research Project, ICRA, and

Dipankor Coondoo is Professor with Economic Research Unit, Indian Statistical

Institute, Kolkata. Corresponding author is Suchismita Bose and her e-mail is

[email protected]* The authors are grateful to Sumon K. Bhaumik of London Business

School, Saumitra Chaudhuri and Mihir Rakshit for their valuable suggestions and

comments.

We make an attempt

to understand the

nature and extent of

imperfection of the

Indian market for

corporate bonds

using data on

secondary market

trading channelled

through the major

stock exchanges.


2/27

26

Money

Finance

of t he spread betw een YTM of d ifferent risk cat egories of bo nds;

relat ionship betw een ma rket depth a nd price/YTM ; a nd t he market

pricing of risk. O urfindings bring out tha t the secondary market for

(exchange trad ed) corpora te bond s is chara cterised by shrinking depth

and w idth in recent years, bo rne out by the decreasing number of tra des

and the rising concentrat ion of tra ding in an increasingly lesser number

of securities. Our sample period w itnesses a q ualita tive change in the

composition of the market, characterised by dw indling of trad e in

highly rat ed bonds and a rise in the trading of dow ngraded bonds,

concentrated in part icular, in bonds w hich have liquidity (irrespective

of their present credit q uality). The usual d istort ionary effects of a

shallow market a re seen to b e manifest in the high volatility of YTM s,

this being more severe for bo nds w ith higher credit risk compared w ith

the ones of h igh credit quality. The observed pa tterns of over time

movement of spread s also suggest tha t in a number of insta nces, private

informa tion/expectat ions w ork strong enough so tha t investors tend t o

ignore the public informa tion cont ained in the declared credit

rat ing of corpora te bonds. Further, our estimates indicate tha t the ma rket

fa ils to evolve a unifo rm ma rket price of risk across rat ing categories.

We observe that much needs to b e done for a meliorat ing the

prob lems of informa tion asymmetry, low liquidity and t he conseq uent

distortions from the corpora te debt segment of t he Indian ca pital market,

to give the debt ma rket a much stronger ba se tha n it currently enjoys.

IntroductionG enerally a dom estic capital ma rket has several segments

viz., commercial ba nks, the eq uity ma rket, non-ba nk financial institu-

tions and the bond market. What should b e the nature of composition of

the capital market for a given economy is largely a policy matter,

although po licies alone canno t determine the compositiona l structure of

the market. In most count ries the debt ma rket segment of the capital

ma rket develops later, a s the financial sector becomes mature. In t he

pre-globa lisat ion era, w hen the para digm of sta te-initiated mixed-

economic development do minated the minds of development econom ists

and policy makers, developing countries like India favoured a capital

market structure having an overwhelmingly large commercial bank

sectormostly pub licly ow ned. The reason fo r such a preference is

rather obvious. Given the approved industrial policy, a relatively large

and centrally controlled banking system would apparently serve as a

pow erful instrument fo r achieving the targeted pat tern of economic

development. The short comings of such a view are too w ell know n

tod ay. An over-sized, mostly publicly ow ned and contro lled ba nking

system wo uld oft en ama ss huge deposits and cha nnel these to preferred

investment plans charging administered interest rates (generally set

below the rate tha t w ould ha ve cleared the ma rket). Such lending

decisions wo uld oft en be taken violating prudent b anking principles.

O perating in an environment of stat e-protection and being heavily

An over-sized,

mostly publicly

owned and

controlled banking

system would often

amass huge

deposits and

channel these to

preferred investment

plans . . . violating

prudent banking

principles.


3/27

27

Money

Finance

leveraged, such a banking system w ould a ccumulate b ad loans, and

more importa ntly, prevent development of other segments of the capita l

market.

As is well known, in a situation where the capital market has

all the three segments, t hree different kinds of funding for investment

may, in principle, be available to business firmsviz., issuance of

equity a nd/or bo nd a nd ba nk lending. N ow, if t he banking system is

largely publicly owned, overreaching, highly regulated and dominates

the other segments, ba nk lending to business firms may f requently go

ba d a nd b ecome unrecoverab le. This is because banks, o perating under

virtual state guarantee, may frequently miscalculate lending risk and

misdirect loans to investment demands tha t ar e not a mong the most

productive ones. In contrast, w hen a w ell developed bond market w ith

a sizeable corporat e bond segment exists alongside the ba nking system,

it is likely that funds will flow in the right direction in accordance with

the productivity of individual investment demands. This w ill be so

essentially fo r tw o reasonsviz., (1) a developed and freely operat ing

corpora te bond ma rket w ill judge the intrinsic wo rth of investment

demand s better in view of the disciplina ry role of f ree market forces

and (2) the corpora te bond ma rket w ill exert a competitive pressure on

commercial ba nks in the mat ter of lending to privat e business and t hus

help improve the efficiency of ca pital ma rket and the econo my a s a

whole.

In fa ct, it is often argued that the Asian financial crisis of 1997,

w hich w as largely a manifestat ion of the w eakness of the financial

system in the countr ies concerned, w as a ccentua ted by t he fragility of

the ba nking system of t he affected countries. The banks, w ith their loa d

of excessive business lending in the form of bot h short t erm w orking

capital loa ns and long term fixed capital loans that w ent ba d a s

borr ow ing firms collapsed due to t he impact of currency crisis, them-

selves collapsed. H ad there been a w ell functioning corpora te bond

market, neither business firms could ha ve ama ssed debt liabilities tha t

turned out to be disproportiona te relat ive to t heir risk adjusted a sset

positions nor the ba nks would ha ve become vulnerable to the shocks

their borrow ers got exposed to (Ra kshit, 2000; see also Herring and

C hatusripitak, 2000 and C orsetti et al, 1998).

It is in the above perspective tha t a n examina tion of the nat ure

of the Indian ca pital ma rket and its different segments assumes impor-

tance as day by day the countrys economy opens up through a variety

of reforms a nd a djustments, thus getting more and more exposed to

external shocks. Ea rlier, w e looked into issues relating to the govern-

ment bond segment of the Indian bo nd ma rket (Bose et al, 1999; Bose

and C oondoo , 1999 and Bha umik et al, 2003). H ere, w e shall try t o

ta ke a glimpse of t he corpora te bond segment of the countrys bond

market a nd d iscuss the issues relat ing to the development of a w ell

functioning corpora te bond ma rket in the country. The next tw o

sections describe the G loba l and t he Indian scenarios w ith respect to the

A developed and

freely operating

corporate bond

market will judge

the intrinsic worth

of investment

demands better in

view of the

disciplinary role of

free market forces.


4/27

28

Money

Finance

corpora te bond market. This is follow ed by t he results of o ur empirical

investigations trying to q uant ify the imperfections in the seconda ry

market for Indian corporat e bonds.

Corporate Bond Market: The Global ScenarioFor business investments debt capital is generally considered to

be more suitab le for large-scale, long-term fina ncing of fixed a ssets a nd

investments, whereas bank loa ns are though t to be more appropriat e for

fina ncing short-term investments in working ca pital, inventories and

other current assets.1Equity ca pital is usually costlier tha n debt, a s

investors w ould expect a risk premium a nd hence a higher return on

equity investment over that f rom a com para ble debt investment.2

However, although debt financing may be an efficient means, most

countries do no t ha ve a w ell developed corpora te bond ma rket. [Box 1

presents an overview of t he role of t he corpora te bond ma rket in a

developing economy and the environment necessary for its develop-

ment.] The only country tha t ha s such a ma rket is the United Stat es.

C ompany financing from bonds for non-financial corporations in the

USA is about 50 per cent, w hereas t he same in Ita ly, is only 2 per cent

(M cG ee, 1998). In contra st, ba nk lending as percentage of G D P f or

Ja pan in 1995 wa s abo ut three times that f or the USA (Sapsford, 1997).

Table 1 gives the structure of the capita l market in the late

1990s of selected developed and developing countries, including India.

As this Table suggests, if the relat ive size of the corpora te bond ma rket

1 Since deposits w ith t he ban ks are mainly shor t/medium term, extending

term loans becomes relatively risky.2 Also, interest expenses on debt are usually tax-deductible at a corporate

level, whereas corporate profits are usually taxed before dividends on shares are

retained or distributed to shareholders. (See Endo, 2000).

TABLE 1

Equities, Government Bonds and Corporate Bonds as percentage of GDP, in

Selected Developing and Developed Countries, late 1990s

Share of Cor porate Bonds (% ) in

Cou nt r y E qui ti es Bank Cl aims Tot al aggr egat e of Equi ti es, Tot al bonds

on Pr ivate Bonds Bank Claims and

Sect or Total Bonds

U SA 158.0 63.6 164.2 18.20 42.75

Ja pa n 65.9 133.3 137.7 11.87 29.05

G erma ny 51.5 125.9 94.5 19.75 56.83

U K 169.7 120.8 61.0 7.91 45.57

Korea 35.7 82.5 86.5 15.10 35.72

India 28.3 26.6 29.3 5.23 15.02

Tha ila nd 30.9 127.8 18.6 1.75 16.67

H unga ry 29.2 17.2 25.1 0.70 1.99

Pola nd 13.0 19.5 8.0 8.00 0.00

Source: Endo, 2000, World Bank.

Debt capital is

generally

considered to be

more suitable for

large-scale, long-

term financing of

fixed assets and

investments,

whereas bank loans

are thought to be

more appropriate for

financing short-term

investments . . .


5/27

29

Money

Finance

is measured in terms of value of outsta nding corpora te bonds in the

aggregate value of equity market capitalisation, bank lending to

private sector a nd a ll outsta nding bonds (i.e., government a nd corpo-

rat e bonds combined), then among the countries considered, t he share is

ab ove 10 per cent only for the USA, G ermany, Korea a nd Ja pan. This

share is a little over 5 per cent for India . If, o n the other hand , the share

BOX 1 : Roles of the Corporate Bond Ma rket and the

Environment Required to Foster its De velopment

A corporate bond market is generally expected to play the following roles:

Diffusing stresses on the banking sector by diversifying credit risks across

the economy;

Diffusing foreign exchange, interest rate and refunding risk; Supplying long-term funds for long-term investment needs;

Supplying long-term investment products for long-term investors;

Lowering funding costs by avoiding a liquidity premium;

Providing products with flexibility to meet the specific needs of investors and

borrowers;

Allocating capital more efficiently;

Reducing reliance on foreign funds, the flow of which can be highly volatile.

The environment conducive to the development of an efficient bond market consists

of:

Macroeconomic and political stability, with a stable and conducive policy

environment, relatively low and stable interest rates;

Good savings rate and limited crowding out;

Tax policies that do not disadvantage bonds;

Supportive legal environment; securities laws, bankruptcy codes etc and

committed regulators;

Developed Government securities market which provides market infrastruc-

ture and supports a profitable, skilled dealer community and provides the

benchmark yield curve for default risk-free securities;

Developed equity market culture;

Infrastructure including, cost effective trading, clearing and settlement sys-

tems;

Market participants, i.e., issuers of size and quantity and institutional inves-

tors who see the mutual benefits.

Note: Studies have found some direct relationships between the relative size of the corporate bond

market in a given country and the relative size of its equity, bank loan, or government bond

market. The latter three markets are generally instituted ahead of a corporate bond market, in

developed countries and their correlation is seen to be strong. The leading role of the govern-

ment bond market is more obvious as a well functioning government bond market helps facili-

tate the growth and functioning of a corporate bond market, in part by establishing a benchmark

yield curve for pricing fixed-income instruments like bonds. The relationship between the bank

loan and corporate bond markets in developed countries is found to be inverted, while that in

developing countries is direct. This is suggestive of the corporate bonds potential to substitute

bank loans on the longer end of the yield curve and thus the ability of a corporate bond market

to relieve the burdens of banking systems in developing countries (Endo, 2000).

. . . value of

outstanding

corporate bonds in

the aggregate value

of equity market

capitalisation, bank

lending to private

sector and all

outstanding bonds

. . . is a little over

5% for India.


6/27

30

Money

Finance

of corpo rat e bonds in the value of all outsta nding bo nds is considered,

this share is much higher for G ermany, the UK, the USA, and Korea

compa red with those of the other countries, the share for India being

a round 15 per cent. It ma y, ho w ever, be not ed tha t t hese shar es need

not reflect the importa nce of the effective corpora te bond mar ket in

these countries as in ma ny count ries the size of the seconda ry ma rket

for corporat e bonds may be rather small.

M any researchers have discussed t he issue of corpora te bond

ma rket development in t he Asian countries (part icularly in Southeast

Asia in the post-Asian Crisis period) and noted the relatively small size

of t he corpora te bond ma rket and its sluggish grow th in these countries.

The main reason put f orw ard by them is the peculiarity of t he financing

pa tterns of b usiness firms in most o f t hese count ries. To be specific, it is

pointed o ut t hat fa mily ba sed co rpora tions/business conglomerates in

Thailand, M alaysia a nd Indonesia tended to prefer a combination of

internal earnings and bank borrow ing to b ond issuance for financing

their fixed capita l investments primarily due to their close and inter-

locking links w ith ba nks and the government. Thus, it is the institu-

tional setting tha t w orks aga inst t he development o f a w ell functioning

corpora te bond ma rket in these countries (Sharma , 2000).

The Indian Corporate Debt Market: A Brief DescriptionD uring the initial phase of expansion of private industry in

post-Independence India, mostly the government-nurtured development

fina ncial institutions (D FIs) supplied long t erm finance to private

industries through various types of financial incentives and supportive

measures. The commercial ba nks w ere na tura lly not keen on providing

such loa ns for f ear of asset-liability mismatch. Working capita l finance,

however, was provided mostly by commercial banks in a regime of

ad ministered int erest-rat e w ith a differentiated ra te structure. This

pat tern of financing changed completely w ith the sta rt of the deregula-

tion process in the 1990s. The D FIs increasingly w ithdrew from project

lending. Their w ithdra w al created a va cuum and t hus the need fo r open-

ing alternat ive sources of term finance to industry and inf rastructure

development came to t he forefront. At present, effo rts are being made

to f ill up this vacuum by enlarging the scope of the bond ma rket and

more particularly encouraging the grow th of an active bond market.3

In pursuit o f o vercoming the financial crisis of t he early 1990s

and evolving a well functioning capital market in the country, a

number of reform measures w ere introduced during t he last decade.4

These have improved the functioning of the eq uity ma rket and the

ma rket for government securities a lot. H ow ever, t hese improvements

3 See for example, Patil, 2000; The Indi a Inf rastructur e Report, June 1996.4 These included several reforms in the securities markete.g., the

establishment and empowerment of the Securities and Exchange Board of India

(SEBI), market determined allocation of resources, nation-wide screen based trading

. . . with the start of

the deregulation

process in the

1990s, the DFIs

increasingly

withdrew from

project lending.

Thus the need for

opening alternative

sources of term

finance to industry

came to the

forefront.


7/27

31

Money

Finance

BOX 2 : Risks Associated with Corporate Debt Se curities

In the market for corporate securities investment decisions are based on the informa-

tion on risk associated with the securities, the investors requirements with respect to

cash flow and their risk return preferences. The risks associated with investment in a

corporate bond are as follows:

Interest rate risk, that is, the risk that interest rates will rise in the future and

thereby reduce a bonds price. This leads to chances of capital loss to the

investor if (s)he wants to sell off the bond at a time when market interest

rates are higher (prices lower).

Reinvestment risk refers to the probability of a fall in market interest rates in

the future, which would in turn reduce the additional income from reinvest-

ing the cash flows from a bond (at a lower rate).

Inflation risk reduces the effective income from an investment in terms of purcha-

sing power. All these risks are obviously higher over a longer holding period.

Liquidity risk, or the degree of accessibility to cash prior to maturity.

Default risk or credit risk refers to the risk that the issuer of the bond may be

unable to make timely (principal and) interest payments on the issue. Default

risk is gauged by quality ratings assigned by the credit rating agencies. The

higher the default risk of an issue, the higher is the risk premium, and hence

yield, associated with it. Characteristics of prices/yields of bonds are ex-

pected to vary according to different degrees of default risk. Thus for the

corporate bond market the analysis focuses on the bonds price/yield behav-

iour in relation to their credit ratings.

notw ithstand ing, the equity ma rket ha s not yet become attra ctive to the

majority of private investors b ecause of the inherent high risk involved

in equity market investment. The debt market (even for government

bonds) ha s also not become popular yet a s a destinat ion of savings of

individual savers either. This has been so essentially because of the

ab sence of a n active secondary ma rket for debt instruments, wh ich

makes investors f eel tha t their investment in debt is highly illiquid.5

in securities, de-materialisation and electronic transfer of securities, rolling settle-

ment and ban on deferral products, sophisticated risk management and introduction

of derivatives trading. All these have helped improve the regulatory framework and

efficiency of trading and settlement. The equity segment of the Indian capital

market is now comparable with that of many developed markets in terms of a

number of qualitative parameters. For developing a deep and liquid secondarymarkets in government securities, the Reserve Bank of India (RBI) has initiated

legal, regulatory and taxation reform, infrastructure and technology improvement,

safe settlement systems, and market dissemination of information on all trades in the

wholesale market. It has also improved methods of issuance of securities such as re-

openings and price-based auctions that allow fungibility of securities, introduced

derivatives such as interest rate swaps, enlarged the repo markets and ensured

adequate liquidity in the secondary markets through the dealers.5 The main finan cial instruments popula r w ith priva te investors/house-

holds are bank deposits, provident funds, insurance, income-oriented mutual funds

and postal savings schemes [see NSE, 2001]. Liquidity generally refers to the ability

to get ou tw ithout moving t he price against the seller. This is only po ssible in a

market where trading is frequent.

The debt market has

also not become

popular yet as a

destination of

savings of

individual savers

essentially because

of the absence of an

active secondary

market for debt

instruments.


8/27

32

Money

Finance

Coming to the corpora te bond segment of t he debt ma rket, such

a market ha s been in existence since Independence in 1947. Public

limited compa nies have b een ra ising capita l by issuing term debt secu-

rities since then mostly t hrough private placement. [Box 2explains the

risks associated w ith corpora te debt securities; Box 3 briefly describes

some features of corpora te bonds; Box 4describes the implicat ions of

credit ratings and Box 5 gives informat ion on priva te placements of

bond s.] From 1985-86, follow ing some debt ma rket reforms, state-

ow ned public sector units (PSUs) began issuing PSU bonds. H ow ever, in

the absence of a w ell functioning seconda ry ma rket, such debt instru-

ments remained highly illiquid a nd unpo pular a mong the investing

populat ion at la rge. As things stand t oda y, the participants in the

Indian debt ma rket a re the Central G overnment, Stat e G overnments,

PSUs, Corpora tes, and Ba nks on the issuing side, and the RBI, commer-

cial banks, insurance companies, mutua l funds, other non-ba nk

financial companies, corpora te treasuries and individua ls on t he

Box 3: Fea tures of Corporate Debt Se curities

Corporate bonds are issuers IOUs and rank senior to both common and preferred

stocks in a corporations capital structure. As creditors, bondholders receive priority

status over the subordinate status of owners, or the corporations stockholders.

Corporate bonds are a direct obligation of the issuing corporations and, at times, are

additionally secured by a lien on specific property, plant or equipment.

In most instances, corporate bonds offer semiannual, fixed interest pay-

ments till a final stated maturity date. Certain bonds also have early retirement

features, known ascalloptions* that allow issuers to redeem bonds prior to maturity

at a specified time and price**. As an alternative, certain bonds are putable, giving

bondholders an option to redeem their bonds prior to maturity. There are other

instruments with embedded options, whose structure and returns are based on a mix

between debt and equity attributes. These include (partially orfully) convertible

debentures or debt instruments that are convertible into equity shares and (share)

warrantswhich are options to buy a specified number of the issuing firms shares, at

a specified price, over a given period of time.

Worldwide, the majority of transactions in corporate bonds is conductedin the over-the-counter (OTC) market, while certain bonds are also listed on the

Stock Exchanges. There are several yieldfigures to consider when evaluating a bond

offering; these different yields take into consideration the coupon rate, the purchase

price and the number of years to a bonds maturity or call date. Yield to maturity

(YTM) represents the return an investor will receive if the bond is held to term. Yield

to call (YTC) is the return earned if a bond is called prior to maturity.

Note: * Optionsare contracts which given the holder the right, but not an obligation, to buy (sell) a

particular financial instrument from (to) the option writer, both the price and time for buying (or

selling) are specified in the contract.

** Callable bonds have a greater chance of being redeemed in a lower interest rate environment,

thus creating reinvestment risk for bondholders. Consequently, callable bonds tend to providehigher rates of return than non-callable bonds. Other bonds have special call features that are

triggered upon the occurrence of an extraordinary event.

With the abolition of

the office of the

Controller of CapitalIssues and removal

of the administrative

control over the

pricing of new

issues, corporate

debt issues

expanded both in

number and variety.


9/27

33

Money

Finance

investment6 side. With the ab olition of t he office of t he Con troller of

C apita l Issues (CC I) and the conseq uent remova l of the administrat ive

contro l over the pricing o f new issues, corpora te debt issues expa nded

to some extent both in number and variety. H ow ever, corpora te issuers

Box 4: De fault Risk and Credit Rating

Credit analysis involves analyses of information on companies and their bond issues

in order to estimate the ability of the issuer to live up to its future contractual

obligationsthe conclusions are given in the form of ratings. Ratings take into

consideration factors like the likelihood of default (particularly under adverse cir-

cumstances), provisions of the debt obligation, protection offered by and the relativeposition of the debt obligation in case of bankruptcy, etc.

Though various rating agencies use different symbols, generally:

Tripe-A denotes the highest safety category

Double A denotes high safety

Single-A denotes adequate safety

Triple B represents moderate safety

Double B is inadequate safety

Single B denote risk prone

C denotes substantial risk and

D denotes defaulted paper

Except for triple A, in other categories use of +orsuffixes indicate above

average or below average credit quality within that letter grade. Paper with ratings up

to BBBis considered to be investment grade, while those below that are regarded

as non investment-grade or junk-bond quality.

A credit rating system is an essential component of any well-functioning

corporate bond market, as it encourages the most efficient allocation of capital raised

by debt issues. Such a system (i) augments the quality and quantity of information on

issuers, (ii) provides the measurement of the relative risk of bonds in question, (iii)

provides bond issuers an incentive for financial improvements, and (iv) alleviates a

loss of liquidity due to security fragmentation. Thus, a credit rating system essen-

tially facilitates the transferability of corporate bonds. Investors will demand ahigher interest rate, commonly known as a risk premium, to compensate for the

higher credit risk implied by a lower rating; this differentiation of interest rates on the

basis of risk in turn helps ensure the efficient allocation of resources by investors

while further encouraging companies to improve their financial performance. A well

functioning credit rating system also encourages greater transparency, increased

information flows, and improved accounting and auditing practices. In addition, the

limited number of creditworthiness symbols alleviates issuer based fragmentation of

bonds and allows for the bundling of bond issues of the same or very similar

creditworthiness into a single category from among the universe of issues. This

creates the ground for interchangeability of bond issues by different issuers and

facilitates arbitrage activities, which in turn can make the bond market more liquid.

6As issuers, corporates issue instruments, namely bonds, debentures and

commercial papers with maturity period varying from 15 days to 1 year (for

commercial papers) and 1 to 12 years (for bonds and debentures).

Dominance of

private placement is

attributed to severalfactorsviz.,

involved issuance

procedure for public

issues, considerably

higher cost of public

issues, much higher

subscription in case

of private

placements etc.


10/27

34

Money

Finance

continued to prefer privat e placement to public issues. Such a domi-na nce of private placement is att ributed to several fa ctorsviz.,

involved issuance procedure for public issues, considerably higher cost

of public issues, much higher subscription in case of private placements

etc. Essentially for t hese reasons, f inancial institutions ha ve tended t o

do minate public issues in the primary corpo rat e debt ma rket. The

seconda ry ma rket for non-sovereign debt (particularly the corpora te

debt segment) also has a number of short comingslack of m arket

making in these securities resulting in poor liquidity, tendency on the

part of institutiona l investors to ho ld these securities to ma turity a nd

thereby reduce market supply of these securities, etc. Tabl es 2 & 3givethe composition of the Indian d ebt ma rket in respect of resource mobili-

sation and issue size in the w holesale debt ma rket (WD M ), respectively.

Both the Tables clearly ind icate the relative unimporta nce of the

corporat e segment.

As already mentioned, not only is the size of t he corpora te

segment of t he India n capita l market small compa red w ith the size of

the government segment, but the seconda ry ma rket for corpora te bonds

is also extremely thin and shallow with very little participation of

individual investors. H ow ever, th is is not an exclusive feature of t he

Box 5: Private Placem ent

The convenience of structuring of issues to match the needs of issuers with those of

investors coupled with savings in terms of time and cost has contributed to rapid

growth of the market for private placement. The rationale for investing in the private

placement market lies in the convenience and flexibility to the issuers as well as

investors.This route is generally preferred by corporates wishing to issue securities

with complex or non-standard features, as deals can be tailor-made to suit the

requirements of both issuer and investor. Many companies may prefer private place-

ments if they wish to raise funds quickly to take advantage of interest rate change in

volatile market conditions. This market is also preferred by new entrants who do not

have track record of performance and hence are unsure about generating adequate

public response for their public issues. Again corporates may prefer this route if the

general market environment is not conducive for floating public issues.

The investors also have advantages in subscribing to private placements,

particularly, when there is no adequate supply of good public issues to match the

amount of investible funds available, investors look for bonds at attractive rates in theprivate placement market. Further, the private placement market provides investors

with securities with more or less fixed/predictable cash inflows, which help the

investor to match the expected stream of returns with the expected cash outflows.

No regulatory compliance is another important reason why corporate is-

suers prefer this route and avoid public issues. In the private placement market, it is

not mandatory to obtain rating on debt instruments, even though some issues are

accompanied by rating. The issuer is also not required to make fair disclosure of all

the credit ratings obtained.

. . . lack of market

making results in

poor liquidity, andtendency on the part

of institutional

investors to hold

these securities to

maturity reduce

market supply.


11/27

35

Money

Finance

Indian corpo rat e bond ma rket. Even in most developed countries, the

seconda ry market fo r corporat e bonds, except for a limited number of

major issues, is generally illiquid, a nd institutiona l investors, ra ther

tha n individuals, a re the key players in these markets. M oreover, o ver-

the-count er (O TC ) and/or the telephonic market, ra ther tha n exchanges,

is in fact extensively used by these institutiona l investors a nd profes

sional money mana gers in most countries. H ow ever, as the major corpo-

rat e issuers in these countries enjoy a fa irly liquid seconda ry ma rket for

their bo nds (such tha t t he available market price data are ad equate forconstructing a benchmark yield curve), bo th ma jor and minor issuers

can ha ve the benefit of a w ell functioning corporate bond market.7In

7 The buy and hold strategy for corporate debt securities, which is often

quoted as a major cause of the illiquidity of their secondary market, is generally

legitimate strategy for institutional investors, who buy in large lots and hold the

securities to maturity, as a substantial part of an institutional investors investment

portfolio does not need to be kept liquid all the time. With the coexistence of both

major and minor issuers bonds, these investors may find it profitable to trade in the

more liquid bonds and hold the less liquid ones. (See for example, Endo, 2000.)

TABLE 3

Composition of Indian Debt market: WDM Segment

I ssue Size

Type of Secur i ty N o. of Secur i t ies (nearest mil l ion)

as on 30 Ja nuary, 2003

G ovt. (C entra l & Sta te) Bonds 685 6046086

PSU Bonds 453 379026

Bank/Financia l Inst itut ion Bonds 420 291327

C orpora te Bonds 257 164845

Tota l* 1815 6881284

* Exclud es Treasury Bills, Certificat e of Deposits, C ommercial Pa pers,

and other Debt

Source: Indian Fixed Income Securities Market, January 2003, Vol. 6, No. 1, NSE

Ltd. Website

TABLE 2

Resources Raised from the Debt Market excluding Treasury Bills

(Rs. Thousand Cror es)

I ssuer 1999-00 2000-01 2001-02

Corpora te

Public Issue 4.5 3.3 6.1

Priva te Placement 61.3 67.8 64.9

Tota l 65.8 71.1 71.0

G overnment 113.3 128.5 132.9

Tota l 179.1 199.6 203.9

Source: M ohamma d Tahir, Development of Bond Ma rket in India,

www.rbi.org.in

In India, the lack of

credit rating

requirements forprivate placement of

corporate debt has

fostered a primary

market structure

where private

placement

dominates in an

overwhelming

manner . . .


12/27

36

Money

Finance

India, the lack of credit ra ting requirements for private placement of

corpora te debt ha s fostered a prima ry market structure w here private

placement domina tes in an overw helming manner, leading to further

lack of liquidity in the secondary market. The privat e placement of

debt a s w ell as tra nsactions in debt securities are generally ma de

through o paq ue negotiations w ith poor disclosures and ineffective

aud its, thus resulting in an inefficient seconda ry ma rket w hich suffers

from f ragmentat ion, low liquidity a nd inefficient price discovery.8

Market Imperfections: Some QuantificationWe made an a ttempt to und ersta nd the nat ure and extent of

imperfection of the Indian seconda ry market fo r corpora te bonds by

examining the fo llow ing aspects: (1) depth of the market, (2) composi-

tion of the ma rket, (3) relat ionship betw een y ield-to-maturity (YTM )

and vola tility of return, (4) nat ure of t he spread betw een YTM s of

different risk categories of bond s, (5) relationship betw een ma rket depth

and price/YTM a nd (6) market pr icing of risk.As regards the da ta

source, we have used available data on tha t fra ction of trad ing in

corporate bonds that are routed through the Bombay Stock Exchange

(BSE) and the Ca pital Ma rket (CM ) and WD M segments of the Na -

tional Stock Exchange (NSE). Naturally, given the lack of width of the

market itself, w e could not have a large data set f or our a nalysis.9

In our a nalysis, w e used ava ilable daily trad ing information

for t he corpora te debt segment of bo th the BSE and the NSE.10This

dailyda ta set covers a 38-mont h period from April 1997 to M arch

2001 (w ith some gaps due to non-ava ilab ility of da ta). O nly non-

convertible debentures (NC D s)11 (i.e., debt securities with no equity

component linked t o t hem) w ere considered. Further, the trad ing da ta

8See IOSCO, 2002, for a comparison of the market microstructure of the

Indian corporate debt market vis-a-vis other emerging markets. For a discussion on

similar problems in the Canadian corporate debt market see Powley (2000).9 It may be mentioned that there is no unique and reliable source of

information on all transactions of corporate bonds. For government bonds, even

though not all secondary market transactions are routed through the WDM of the

NSE, all trades are recorded with the RBI and therefore one gets the entire data

from the RBI sources. However, in the case of corporate bonds a significant portion

of the trade is done as bilateral agreements between two counter-parties, informa-tion on which is not readily available. Also, for our purpose, data on exchange-

traded transactions was of greater relevance, since it is the exchanges that provide

the basis for efficient market making by reducing information asymmetry. Hence,

we focussed on that fraction of trading in corporate bonds that were routed through

the BSE and the Capital Market (CM) and WDM segments of the NSE.10 The data were collected from daily trade papers.11 In the USA, debentures are defined as debt instruments which are not

secured by any pledge of assets (either fixed assets or financial obligations of other

companies). In the UK, on the other hand, debentures usually refer to secured debt.

In India, debentures are defined as debenture stock, bo nds and a ny oth er securities

of a company, w hether constituting a charge on the assets of the company or not ,

under Section 2(12) of the Companies Act, 1956.

We used available

daily trading

information for the

corporate debt

segment of both the

BSE and the NSE

from April 1997 to

March 2001.


13/27

37

Money

Finance

for bonds issued by financial institutions (FIs) w ere excluded.12

It should be pointed out here tha t our cho sen sample period,

unfort unat ely, w as a period w hen several external shocks struck the

economy a nd the industry suffered a protracted period of slow ing dow n

of grow th. A major disturbance, one that had a direct impact on b ond

markets, w as the w eakening of the Indian rupee and the conseq uent

policy measures of the RBI to counter this w eakening. Specifically, t he

RB I reversed its easy money policy and a s a result the interest ra te

increased significantly.13At the same time, w ith a slowd ow n of eco-

nomic growt h, it w as strongly anticipated tha t the high interest rate

policy itself w ould b e reversed as soon as t he rupee had been able to

consolidat e its position vis--v isthe US dollar at a new equilibrium

level. Further, since there wa s no obvious indicat ion a s to ho w soon

such stabilisation might occur, given the experience of the Southeast

Asian countries, there w as a pprehension abo ut the future of w eaker

compan ies (tha t ha d fa ced a rise in interest cost and a po ssible

slow dow n in econom ic activities simulta neously). Thus, the interest

rat e condition w as strong ly influenced by short-term developments on

the foreign exchange front till the financial year 1998-99. The interest

rat e in the Indian economy moved dow nw ard d uring 1999-2000

follow ing the reversal of t he RBIs moneta ry policy in M arch 1999.

Signs of industrial recovery, improved corpo rat e sector performa nce

and strengthened ma cro econo mic funda mentals (such as, low inflation

along w ith a sha rp rise in the prices of informa tion technology stocks

w orld-w ide) led t o a susta ined rise in share t rad ing during 1999-00,

w hich continued unt il the bursting of the asset price bubble affected

financial markets a ll over the w orld. These events set t he ba ckground

for t he present ana lysis of the nature of the seconda ry corpora te bond

segment of the Indian capita l market.

As regards the health of the secondary ma rket for corpora te

bonds during the period of o ur study, as our da ta set indicates, frequency

of t rad ing in corporat e bonds (on exchanges) declined fro m Februa ry

1998 onw ard s. This declining t rend continued till the end of our sa mple

period w ith o ccasional short reversals (e.g., d uring the September-

D ecember 1998 period). Trad ing in corpo rat e debenture almost halved

during 1999 and o ur dat a show ed a dra stic decline from N ovember 1999

onw ards. It can be seen from Table 4tha t the tra ding in non-government

bonds as a w hole declined by more tha n half f rom its April 1997Ma rch

1998 level during the next tw o years. This w as follow ed by ano ther

12 This was done for the following reasons: First of all, FIs are perceived to

be quasi-government institutions with significantly low default-risk. Therefore,

unlike the bonds issued by corporate entities, bonds issued by FIs are highly traded.

Further, in recent years most bonds issued by FIs are hybrid instruments with a

variety of features like embedded options, and therefore it is difficult to compare

these bonds with NCDs issued by corporates.13 For details about the RBIs direct and indirect interventions in the

market for foreign currencies, see Mukhopadhyay (1999).

A major

disturbance, one

that had a direct

impact on bond

markets, was the

weakening of the

Indian rupee and the

consequent policy

measures of the RBI

to counter this

weakening.


14/27

38

Money

Finance

similar sha rp decline during April 2000M arch 2001, w hich could

partly be due to crowding out effecta s these periods w ere also periods

of increasing volumes in the gilt market and the equity ma rket.14The

bo nds of f inancial institutions were heavily trad ed betw een April 1998

and Februa ry 1999, w hich could have accounted for pa rt of t he reduced

trad ing of corporate bonds during that period. Table 4also show s that

there w as a significant ga p betw een the number of tra nsactions in non-

government bo nds and the number of trades in FI and corporate bonds

put to gether. There w ere a number of o utsta nding corpora te debt

instruments, w hich w ere hybrid in nat ure, combining features of bot h

debt a nd equity (convertible debentures and w arra nts); these accounted

for most of the trading d uring the latter period,15w hile there wa s a

decline in t rad ing in the FI bo nds a s w ell as no n-convertible debentures.

In w hat follow s, w e summarise the various results that w ere obta ined.

Depth of the MarketIn order to verify the extent o f depth (or lack of depth, to be

specific) of t he secondary co rpora te bond ma rket, w e tra cked the n-

bo nd concentrat ion rat io for freq uency of t rad ing over time. To be

precise, fo r each mont h w e identified a set of nmost frequently traded

corpora te bonds a nd measured their share in the tota l frequency of

trad ing during tha t month. C learly, if the n-bond concentration ratio

turns out to be high for a small value of n,one may safely conclude

that the market lacks depth, because in that case the probability will be

low that a b ond tha t does not belong to the chosen set w ill be traded.

Since it is the frequency of t rad ing of b onds that determines the extent

of m arket efficiency associated w ith determination o f their prices,16a

high n-bond concentra tion ra tio should, in turn, imply presence of a

large number of perverse prices in the secondary market. In our exer-

cise, w e used t hree n-bond concentration rat ios for n = 5, 10 and 15,

respectively.

As the va lues of the three n-bond concentration measures

presented in Table 4suggest, there wa s a ra ther high level of concentra -

tion of trad ing in the secondary corporat e bond ma rket during the

period und er review. M ore importan tly, this concentra tion show ed a

14 The boom in the equity market did not seem to have any positiveimpact on the corporate debt market, at least not on secondary market trading.

C ampb ell and Taksler (2002), show tha t dur ing the lat e 1990s even th e US equ ity

and corporate bond markets behaved very differently; stock prices rose strongly,

while corporate bonds performed poorly.15 Particularly, a series of bonds from Reliance Petro and Nirma accounted

for the majority of these trades.16 Low frequency of trading implies that (a) trading takes place between

very few players, and (b) there is inadequate signalling of interest rate expectations

prevailing in the market. Hence, players are then more vulnerable to price manipula-

tion. Further, in the absence of adequate information about expected interest rates, it

would be difficult for the investors to converge to a unique set of theoretically

correct prices, and hence price volatility would be exacerbated.

If then-bond

concentration ratio

turns out to be high

for a small value of

n,one may safely

conclude that the

market lacks depth.


15/27

39

Money

Finance

bro ad rising trend o ver the period o f study, reaching extremely high

levels during April 2000 to M arch 2001. It ma y be mentioned tha t a

similar time trend of increasing concentrat ion of tra ding w as also

observed in the gilt ma rket in our study o f the Indian go vernment bo nd

market conducted earlier.17Po ssibly, t he uncerta inties relat ed to the

Asian f inancial crisis show ed up in these ratios. As a closer look a t t he

time series of these concentrat ion measures show s, the concentrat ion

first increased during mid-1997 and af ter coming dow n for a brief

period d uring November 1997 to Ja nuary 1998 by a bout 10 percenta ge

points, aga in w ent up. There w as a jump of a bout 15 percentage points

in these rat ios during September 1998, coinciding w ith the interest ra te

hike effected by the RBI in late August1998. As Table 4 show s, the 15-

bond concentration ratio w ent up in each financial year, from around

70 per cent in 1997-98 to over 76 per cent in April-November 1999.

D uring the 2000-01 period, w hen most of t he trad ing might ha ve shifted

to o ther markets like the gilt or equity ma rkets, the 15-bond concentra-

tion ra tio w as a larmingly high a t a round 94 per cent. The seriousness

of the situa tion is also reflected by t he value of the 5-bond concentra-

tion ra tio, w hich wa s consistently near 50 per cent a nd jumped to 85

per cent in 2000-01.18

We also approa ch the issue of depth in ano ther w ayviz., by

identifying the extent to w hich tra ding is concentrated a mong bo nds

w ith high credit ra ting. For this purpose, w e consider only the fre-

17 Even at present, when the secondary market for gilts has grown

manifold, it is evidenced that at any approximate point in the yield curve only a

single paper may get traded, e.g. the 9.81 per cent 2013 paper was the only paper in

the 10 year tenure range to trade through January, February and part of March,

2003.18 The concentration ratio for the most-traded 10 and 25 gilts also went

up by more than 15 percentage points between 1998-99 and 2000-01, to 58 and 81

per cent, respectively.

TABLE 4

Trade in non-government bonds: Apr97-Mar01

Sub-per iod Average monthly frequency of t rade of Relat ive frequency* * of t rading of

non-govt . I nst i tut ional Corporate 5 most 10 most 15 most

bonds bonds bonds* f requent ly f requent l y f requent ly

t raded cor po- t raded cor po- t raded cor po-

rate bonds rate bonds rate bonds

Apr97 to M ar 98 24209 1253 4895 47.45% 60.86% 69.73%

Apr98 to Feb 99 9128 1746 2755 50.44% 65.57% 73.95%

Apr99 to Nov 99 10163 791 1277 49.59% 67.01% 76.38%

Apr 00 to M ar 01 4240 617 450 84.89% 92.24% 94.37%

Apr 97 to M ar01 12740 1168 2630 58.45% 71.29% 78.31%

Source: D aily tr ade papers

* Refers to NCDs( non-convertible debentures only)

** Frequency of t rad ing of th e 5/10/15 most frequently tra ded bond s, in each mo nth, as a percentage

of total frequency of trading in corporate bonds.


16/27

40

Money

Finance

q uency of tra ding and do no t explore the volume of trad e (i.e., the tota l

value of t rad e). The preference for freq uency of tra ding over the

volume of t rad e is essentially due to our presumption tha t t he efficiency

of the market a nd t he pricing mechan ism is more closely relat ed to t he

former.

Composition of the MarketAs is w ell know n, corpora te bonds issued in India a re rated for

their credit w orthiness. Bond s of a variety o f ra tings are outstand ing at

any point o f time. The rating cat egories sta rt a t the top f rom triple A,

and move do w n thro ugh do uble A, A, triple B a nd no n-investment

grades, starting w ith double B and ending at D . Within each letter

category, differences over and below the average are indicated by

suffixes of + or . All rat ing categories of bonds are equally importa nt

so fa r as seconda ry market tra ding is concerned. In f act, a n indicato r of

the degree of market efficiency wo uld be whether or not , bonds of

various ra tings get tra nsacted regularly in the market. In o ur exercise,

w e exa mined t he month-w ise freq uency of trad ing of b onds of different

ratings among t he 20 most frequently trad ed bonds during a month and

the share of bonds of different rating cat egories in the tota l frequency of

trading.

Charts 1A-1D present the distribution of to ta l trad ing by

categories of bond for f our sub-periodsviz., April 1997 to M arch

1998, April 1998 to February 1999, April to N ovember 1999 and April

2000 to M arch 2001. Chart 2 show s the over time movements of the

share in the tota l trading of bonds of different ra tings for the study

period.19The following picture emerges from these charts:

For April 1997 to M arch 1998 (i.e., the first sub-period of our

38-mont h study period), the share of t riple A and d ouble A rated bo nds

together w as ab out 35 per cent, a nd a s Chart 2 would suggest, this

share show ed a dow nw ard trend. Among AAA rated bonds, NC D s of

Reliance Industries w ere tra ded heavily. Single A rated bonds, on t he

other hand, accounted fo r 24 per cent of this trading and maintained

this share more or less. Bond s of M angalore Refinery and Petrochemi-

cals (M RP L) account ed for bulk of the trad ing in this group.

D uring the second sub-period (that is, April 1998 to Februa ry

1999), the share of AAA and AA rat ed bonds together went dow n to 22

per cent, w hereas the share of A rated bonds went up to 30 per cent,

w ith bonds of M RPL, and D eepak Fertilisers and C hamba l Fertilisers,

contribut ing to the bulk o f t his increase. This sub-period is also ma rked

by a sha rp increase in the share of non-investment grad e bonds (that is,

category ma rked Restin the Chart).

19 From this point onwards we have ignored all papers which do not have

a declared credit rating (for any month). We have also seen that the problem of

multiple ratings from the three rating agencies (namely ICRA, Crisil and Care) does

not occur at any point of time.

An indicator of the

degree of market

efficiency would be

whether or not,

bonds of various

ratings get

transacted regularly

in the market.


17/27

41

Money

Finance

D uring the third sub-period (April to November 1999), w hile

the pooled share of AAA and AA rat ed remained a t 22 per cent, the

share of AAA ra ted bond s fell to 5 per cent f rom it s previous 8 per cent

level and t hat of the non-investment cat egory rose to 42 per cent f rom

the 29 per cent level of the previous sub-period .

Finally, during the last sub-period April 2000 to M arch 2001,

no tra ding of AAA rat ed bonds was ob served. The shares of A and BB B

rat ed securities went do w n further to 15 and 11 per cent, respectively,

and tra ding in non-investment gra de bond s surged. Indeed, the share of

such bonds sw elled to 52 per cent (during a period w hen tota l number

of reported tra des declined sha rply). D uring the last three sub-periods,

trad ing w as concentrated in the Restcategory consisting of BB+ to D

rat ed corpora te papers like Jinda l Vijayna gar Steel, H otel Leela

Venture, Torrent G ujara t Bio tech, Timex Watches, Essar O il and

H industhan Development Co rporat ion, among others.

CHART 1A

Type of Bonds Traded During Apr97 to Mar98

CHART 1B

Type of Bonds Traded During Apr98 to Feb99

CHART 1D

Type of Bonds Traded During Apr00 to Mar01

CHART 1C

Type of Bonds Traded During Apr99 to Nov99

No te: By a ny ra ting group X we mean ratings X+ ,X and X-. Rating changes have been counted separately if

the rating group changes, i.e., a rating change from say BBB to D within a month has been recorded both

under BBB group and Rest group; the last rating has been counted if the change is within a group. From

this point onwards we exclude any bonds which do not have a declared credit rating.


18/27

42

Money

Finance

O n the w hole, thus, the period w itnessed a qua litat ive change

so far as the composition of the market w as concernedviz., a dw in-

dling of trade in highly rated bonds and a sharp rise in the trading of

non-investment grade bond s. The first phenomenon ma y be interpreted

as a n expression of market pa rticipants strong preference for security

(as the holders of h ighly rated corpo rat e bonds evidently preferred no t

to part w ith these). The second phenomenon ma y a pparently be sugges-

tive of t he emergence of a junk bond market in a regime w here interest

rate has been moving dow nw ard. In fa ct, tow ards the end of our study

period, the share of these bo nds soa red to t he 75-80 per cent level.

H ow ever, a closer scrutiny indicated tha t these bond s were not t ypi-

cally high risk/return (junk) bond s, but w ere ra ther medium risk bond s

(dow ngraded ones) that mana ged to retain their liquidity in the thin-

ning market. O ne may regard such predominance of trad ing in dow n-

graded bo nds as a n indication of a specific type of portfo lio ma nage-

ment technique ad opted by some institutiona l investors, w ho, given

their w eak ba lance sheets, might ha ve been trying o ut a high risk-

return strategy, leading to a surge in trading of high yielding bonds.

H ow ever, this phenomenon could also be an indication o f the inability

of investors to access higher rat ed bonds as, in a situa tion of extreme

econo mic uncerta inty, investors ma y be holding o n to higher rated

bo nds. H ence tho se w ho tra ded to meet their liquidity requirements

might ha ve been forced to tra de in lesser quality bonds.

0%

20%

40%

60%

80%

100%

Apr-

97

Jul-

97

Oct-

97

Jan-

98

Apr-

98

Jul-

98

Oct-

98

Jan-

99

Apr-

99

Jul-

99

Oct-

99

Jan-

00

Apr-

00

Jul-

00

Oct-

00

Jan-

01

AAA AA Grp A Grp BBB Grp Rest

CHART 2

Relative frequency of transaction of 20 most frequently traded debentures; Apr 97-Mar 01.

See note below Chart 1.

YTM and its VolatilityIn contra st to a government bond, w hich involves only interest

rate risk, a corporate bond carries both a default risk and an interest

rat e risk. Bo nd ra ting helps investors d istinguish betw een high risk and

low risk bond s, so to say. If the rating is done correctly and the market


19/27

43

Money

Finance

functions efficiently, it is expected that market prices will move such

that the YTM of a bond o f higher rating w ill be low er than tha t for a

bond o f low er rating (the difference being the risk premium), given tha t

the bonds have the same number of years maturity. In other w ords,

market eff iciency w ould minimise perverse pricing o f b onds tha t might

result in YTM-rating mismatches. We tried to examine the extent of

such mismatches for t he Indian corpo rat e bond ma rket.

G iven the a vailable bond price da ta, w e calculated correspond-

ing YTM s [Box 6gives our method of estimat ion of the YTM s].20The

results suggested t hat on t he w hole the pricing mechanism of t he Indian

corpora te bond ma rket, in spite of its underdeveloped sta te, w as

consistent w ith the theory of bo nd pricing. Thus, for example, it wa s

found t hat for a ll categories of ra ting, YTM of a bond declined a s the

bond approached maturity a nd a lso, given the number of y ears to

maturity, YTM o f a higher rated bond w as lower than tha t of a low er

rated bo nd, on an a verage. H ow ever, lack of d epth of the market

seemed to result in significant va riat ions in the price/YTM o f even AAA

BOX 6

Estimating Yie ld to M aturity

The price data is used to estimate the YTMs using the approximation*

AYTM [(M P) +AC] / [(M+P)/2]

N

Where:

Mis the maturity value;

Pis the price;

Nis the no. of years to maturity ;

ACis the total coupon payments per year.

The spreads are calculated from the reported YTMs of gilts of similar

maturity, averaged over the relevant time period.

Note: * This approximation suggested by Meir Kohn in Financial Institutions and Markets (p.88) gives

estimates that may differ from the actual YTMs by 10-15 basis points, this is a reasonably good

indicative figure for the average annual YTM at least from the point of view of any economic

analysis. Current yield, which is conventionally reported in the debt market, fails to capture

capital gains (losses) arising out of the difference between the issue price and traded price of the

bond, which does matter significantly, particularly if the bond matures in the near future.

20We cannot claim to produce very accurate results on the YTMs and

spreads of corporate bonds as the various redemption features and lack of certain

information on these, make YTM calculations difficult, and the derived YTMs may

not be strictly compara ble across bonds. Neither ha ve all bonds been uniformly tra ded

in all months during the sample period, making comparisons even more difficult.

The usual

distortionary effects

of shallow market

possibly got

manifested in high

volatility of YTMs,

this being more

severe for bonds

with higher credit

risk compared with

the ones of high

credit quality.


20/27

44

Money

Finance

Spread between YTM of Corporate and Government BondsG iven the estimated YTM of bonds of different ra tings, w e

have estimated the spread betw een the YTM of corporate bond and

government bond (i.e., the excess of Y TM of corpo rat e bond over tha t

of go vernment bond) for corpora te bonds of different rating cat egories.

The YTM of go vernment bo nd is alw ays fo und to be smaller tha n tha t

of a corporate bond of comparable maturitythus indicating tha t the

ma rket behaves in such a ma nner that the resulting prices allow the

expected risk premium. H ow ever, the YTM of a low er rated bond is not

alw ays ob served to be systematically larger tha n tha t of a higher ratedbo nd, indicating thereby tha t the market prices of bo nds of different

ratings sometimes fail to yield the expected risk premium. The meas-

ured spread of YTM for bonds w ith less tha n 3 years of residual mat u-

rity and that for bo nds w ith 4 to 5 years of residual maturity are seen

to vary rat her considerab ly over mont hs. This is true even for AAA

rated corporate bond.21Further, w hile the spread of Y TM of , say, BBB

TABLE 5

Price/ Yield Volatility of bonds over rating categories, during Apr 97 Mar 01.

C.V(% ) of prices S.D of YTM s

N o. o f average M inimum M ax imum range average M inimum M ax imum range

secur i t ies C.V C.V S.D S.D

G ilts 6 0.55% 0.02% 0.85% 0.83%

AAA 6 2.49% 1.06% 3.94% 2.88% 1.28% 0.74% 2.46% 1.71%

AA+ 6 2.88% 0.63% 5.58% 4.95% 0.87% 0.40% 1.98% 1.58%

AA 7 2.71% 0.44% 6.22% 5.78% 1.84% 0.57% 4.15% 3.58%

AA- 5 3.54% 1.29% 6.67% 5.38% 1.12% 0.42% 1.81% 1.39%

A+ 7 4.10% 1.67% 8.53% 6.86% 2.03% 0.57% 5.75% 5.18%

A 5 3.37% 1.14% 7.00% 5.86% 1.72% 0.88% 3.16% 2.29%

A- 2 2.67% 2.35% 3.00% 0.64% 1.00% 0.82% 1.18% 0.36%

BBB+ 6(4) 4.12% 1.65% 7.37% 5.72% 2.03% 0.60% 4.18% 3.58%

B B B 7(3) 2.66% 1.62% 3.79% 2.18% 1.62% 0.88% 2.92% 2.04%

BB B- 4 4.35% 2.21% 9.65% 7.43% 2.12% 0.86% 4.51% 3.65%

D 7(3) 11.62% 4.81% 30.67% 25.87% 4.89% 3.48% 6.95% 3.47%

No te: Figures in brackets are the number of securities for which YTM w as estimated; for the rest proper

informat ion on maturity dates and options w ere not ava ilable.

The gilts cover maturities similar to the corporate bonds considered.

and AA rat ed bonds. Volat ility (measured as stand ard deviation of

YTM ) wa s found to be larger for low er rat ed bo nds than f or higher

rat ed bo nds [see Table 5]. In other w ords, the usual distortiona ry effects

of shallow market possibly got manifested in high volatility of YTMs,

this being more severe for bo nds w ith higher credit risk compared w ith

the ones of high credit qua lity.

21 This could happen due to fluctuation in yields of both the corporate

bond and the G overnment bond from w hich the spread is calculated, as these bonds

were close to maturity.


21/27

45

Money

Finance

rat ed bond over AAA rat ed bond , is observed to be positive and large

alw ays, the spread of Y TM of A rated bond over AAA rated bond s w ith

residual maturity of less than 3 years is not always found positive, al-

tho ugh the spread w orks out t o be positive on the average. Such results

are observed fo r bonds w ith greater tha n 3 years to ma turity as w ell,

w hen the spread is calculated for bo nds of a djacent rating cat egories

(i.e., say, between AA and AAA or between A and AA, etc.). 22A closer

look suggests that the impact o f ra ting revision (or even a n expected

one) may be strong enough to cause such aberra tions. Charts 3A & 3B

CHARTS 3A

Spreads of Corporate Securities by Rating with 3-5 years to maturity

CHARTS 3B

Spreads of Corporate Securities by Rating with less that 3 years. to maturity

22 Diaz and Navaroos (2002) analysis of yield spreads between Spanish

Treasury and non-Treasury securities and its relation with term to maturity showed a

downward sloping term structure of yield spreads of investment grade bonds, cont-

rary to theoretical literature, but they found that this outcome was caused mainly

by the effect of liquidity on spreads. Similar indications were also found in our data,

where the spread of a AAA rated bond with less than 3 years to maturity could be

higher than a similarly rated bond with greater than 3 years residual maturity.

. . . at times private

information works

strong enough so

that investors tend

to ignore the public

information

contained in the

declared credit

rating of corporate

bonds.


22/27

46

Money

Finance

show the over time movement of spread of bonds of different ra ting

categories (vis-a-vi sgovernment bond s). It ma y be seen in C hart 3A

that the spread of Y TM of single A rated bond over AA rated bond w as

the consequence of the YTM of A rated bo nd (which had a po sitive

outlook and w as in fact upgraded to the AA group aft er a f ew mont hs)

dipping sharply enough to ca use a negat ive spread w ith the next ra ting

category. In Chart 3B, the A rat ed bond ha s been upgraded one notch

to A+ lead ing to a decline in its YTM , w hich is shar p enough in a

shallow market to cause a lower spread compared with even a AAA

rat ed bond of near similar mat urity. The observed patt erns of o ver time

movement of t hese spreads thus possibly suggest tha t a t t imes private

informa tion (or the belief of t he investors) works strong enough so that

investors tend to igno re the public inform at ion conta ined in the de-

clared credit rating of corpora te bonds.

Whether Depth aff ects Price and YTM

As already mentioned, a major adva ntage of ha ving a w ell

functioning corporate bond market w ith an active seconda ry ma rket

segment in an economy is that such a mar ket ensures cautious evalua-

tion/screening o f t he risk and viability o f pro posed investment pla ns of

enterprises by the ma rket mecha nism itself. In fa ct, in an efficient

corpora te bond mar ket, the price of a bond reflects buyers perceived

yield/return (duly a djusted fo r the risk invo lved) fro m the bo nd, given

their expectation a bo ut the macro econo mic funda mentals. A market

lacking depth w ill nat urally give rise to perverse prices tha t w ill fa il to

reflect these expectat ions accurately and , more importa ntly, w ill not b e

ab le to judge the viability o f pro posed investment plans correctly.

Essentially for this reason, it is wo rthw hile to exa mine w hether or not

the depth o f ma rket af fects the price/YTM of bonds.

To exa mine the extent of perverse pricing of b ond s, w e esti-

ma ted the volat ility o f the price/YTM of b onds of different rat ing

categories and compa red the volat ility a cross rating cat egories. Anec-

do ta l evidence suggests that t he seconda ry market fo r corpora te bonds

becomes progressively t hinner as the default risk of bond increases. If it

w as so, volat ility of price/YTM should decline as ra ting w ent up. We

also bro ught in the stand ard deviation (SD ) of YTM of go vernment

bonds in this comparison. Table 5presents the average, minimum,

maximum and range of coefficient of variation (CV) of price and SD of

corresponding YTM for gilt and 11 rat ing categories of corpo rat e bonds

over the entire study period April 1997 to M arch 2001. For a given

rating category, the CV of price and SD of Y TM for each month w ere

computed using all the observed price dat a for t hat category of bond

ignoring the period to mat urity. The results bro ad ly support the hypot h-

esis of an inverse relationship betw een credit w orth iness (i.e., the

rat ing) and extent of vola tility of price/YTM . It ma y be noted tha t the

average value of CV of price rises as rating goes down. The range of

volatility a lso show s a b road tendency to increase as rating goes dow n.

The results broadly

support the

hypothesis of an

inverse relationship

between credit

worthiness (i.e., the

rating) and extent of

volatility of price/

YTM.


23/27

47

Money

Finance

For SD of YTM , ho w ever, the tendency o f such inverse relationship

w ith rat ing is fa r less prominent.

Market Pricing of RiskAs is w ell know n, holding of a corpora te bond involves bearing

of a composite of defaultriskand returnrisk. The rating of a bond

reflects the default ri skinvolved in holding the bond w hereas the return

riskis governed by t he volatility o f price/YTM of the bo nd. As w e have

already seen, the degree of vo latility of Y TM of b onds of different

rat ings is not the same across rating cat egories. In other wo rds, bond s

of different ra tings carry retur n r iskof d ifferent degrees. No w, if the

market is w ell functioning a nd efficient, it should give rise to prices of

bonds of different ratings (of equa l residual maturity ) such tha t the

implied market price of risk is the same a cross bond categories, w here

the market price of risk for a given cat egory of bond is measured by t he

value of Sharpe Ratio f or this category of bond (viz., [YTM for t he

bond cat egoryrate of riskfree return]/SD o f YTM for t he bond

category ). Tab le 6 presents the estimated va lues of the Sharpe Rat io fo r

different categories of bo nd ra ting based on a vailable price data for t he

period up to end-1999 (as compara ble data for t he remaining period

w ere not a vailab le). We used the YTM of go vernment bonds of compa -

rab le residual ma turity as t he measure of t he risk-free return in these

calculations. As these ava ilable estimates suggest, the ma rket w as q uite

far from being efficient and well functioning as it failed to evolve a

uniform ma rket price of risk across rating ca tegories.

Indeed, it w as rat ed b onds in the BBB category, w ith residual

mat urity less than o r equa l to 3 years, fo r w hich the excess return per

unit of risk borne turned out to be the highest. This w as fo llow ed by

AA+ rat ed bonds w ith residual maturity b etw een 4 and 5 years. These

results also keep open the possibility tha t bo nds belonging to the same

TABLE 6

Sharpe Ratio for Corporate Bonds with different Ratings

Residual M aturity

Rat i ng 4-5 years < 3 years

AAA 2.78

AA+ 5.88AA+ /AAA 3.85

AA+ /AA 4.00

AA 3.33

A/A+ 2.94

AAA/A 2.22

AA+ /A 2.08

A-/BB 3.85

BBB+ /BBB- 6.25

B B B -/D 1.72

No te: These estimates are based on data for the period up to end 1999.

Available estimates

suggest, the market

was quite far from

being efficient and

well functioning as

it failed to evolve a

uniform market

price of risk across

rating categories.


24/27

48

Money

Finance

rating category may have YTM and its volatility such that the implied

Sharpe Ra tio value may va ry from bond to bond.21, 22Interestingly

enough, in our exercise w e got some clear evidences of reduction in

Sharpe Ra tio value follow ing dow ngrading of a bond f rom AAA to AA+

rat ing category or fr om investment to non-investment cat egory, say.

Needless to mention, a ll these are indicative of the immat urity of the

Indian corporate bond ma rket.

ConclusionIn this paper we mad e an at tempt to examine the nature of the

Indian corporate bond market using available data on secondary

ma rket trad ing channelled through the NSE and BSE during the 38

mont hs from April 1997 to M arch 2001, w ith some ga ps in betw een.

O ur primary o bjective w as to judge the extent o f inefficiency of this

market in view of t he w ell-know n fa ct tha t this market is quite thin and

shallow. For this purpose, w e exa mined several aspects of the ma rket

such as, depth and compo sition of the market, relationship betw een

YTM and volat ility o f return a s implied by ob served price movements,

na ture of spread b etw een YTM of d ifferent categories of bond , relat ion-

ship betw een ma rket depth a nd price/YTM and , f inally, ma rket pricing

of risk.

O ur exercise suggests the follow ing qua litative feature of the

Indian corpo rat e bond market, a s it stands now. Like the government

bond market, the seconda ry ma rket fo r corporate bonds to o is marked

by lack of d epth and w idth. H ow ever, a s opposed to the former, w hich

ha s been expand ing, the secondary ma rket for (exchange traded)

corpora te bond ha s been chara cterised by shrinking depth and w idth in

recent years. D espite the problems of measurement of YTM s, and hence

comparison o f t he same betw een bo nds, the picture we ob tained w ould

cause concern about the stat e of the corpora te debt ma rket. The imper-

fections in the secondary ma rket for corpora te bonds a re manifest in

infrequent tra ding, high liquidity risk, a high degree of dispersion of

price/YTM over time, and a lack of strong a nd unidirectional relation-

ship betw een a b ond s credit ra ting (risk) and it s market price/YTM in

23 The divergence of risk perceptions of individual bonds, within a rating

cat egory is how ever, not an uncommon phenomenon. There has been extensive

development of rating based reduced-form models, which take as a premise thatbonds when grouped by ratings are homogeneous with respect to risk, Elton et al

(2002) examine Moodys and Standard & Poors ratings of corporate bonds and

show they are not sufficient metrics for determining spot rate curves and pricing

relationships.24 In our exercise, we did notice cases of widely different Sharpe Ratio

values of bonds belonging to the same rating category. This shows that in a shallow

market risk/return may va ry q uite widely for individual bo nds w ithin a rating

category. Evidence to this effect has been provided by studies like that of Campbell

an d Taksler (2002), w ho used US panel data for t he late 1990s and show ed tha t

idiosyncratic firm-level equity volatility might explain as much cross-sectional

variation in yields as could credit rating, perhaps because equity volatility would

reflect recent information that might not yet be reflected in the credit rating.

Imperfections in the

secondary market

for corporate bonds

are manifest in

infrequent trading,

high liquidity risk, a

high degree of

dispersion of price/

YTM over time, and

a lack of strong and

unidirectional

relationship

between a bonds

credit rating (risk)

and its market price/YTM.


25/27

49

Money

Finance

q uite a number of insta nces. G iven the current slack in overall invest-

ment activity in the Indian economy, the primary and secondary

markets fo r corpora te debt, represented by the private placements

market and an OTC market, may seem to be sufficient. But once the

investment climat e improves and t he demand f or long t erm funds picks

up, the need for a vibrant secondary ma rket fo r corporate debt w ould

definitely b e felt ra ther a cutely.25

G iven its present stat e, the basic questions about the develop-

ment of the Indian (seconda ry) market for corpora te bonds a re (1) how

essential is it tha t the market grow s and becomes a mat ured one and

(2) w hat steps w ill fa cilitat e such a grow th. So fa r as the issue of

essentiality is concerned, as is w ell know n, leaving a side the interna-

tiona l financial ma rket, the three principal sources finance for business

investment are the commercial banks, the equity ma rket and t he

corporate debt market. Of these, a commercial bank loan is typically

appropria te for short-term requirements, w hereas fina nce obtained fro m

the equity o r the debt m arket w ould serve long-term investment require-

ments w ell. Since capita l ma rkets, in developing countries, in part icu-

lar, a re far from being frictionless and efficient, the effective costs of

capital in these two markets are not the same. It is generally agreed

tha t equity f inancing of investment turns out to b e more expensive

because of the associated risk involved. H ow ever, in the a bsence of a

w ell functioning corporat e bond ma rket, particularly in the developing

countries, business investors turn t o commercial ba nks for loa ns. In

such a situation, development of t he corporat e bond ma rket (its second-

ary market segment, in pa rticular) becomes an essential prerequisite for

efficient f inancing.

As Ha kansson (1999) puts it, w hen the relat ive sizes of t he

banking system a nd the corporate bond market are more balanced, a s

w ould be the case when a w ell developed corpora te bond ma rket is

present, ma rket forces ha ve a much greater opport unity to assert

themselves, thereby reducing systemic risk a nd proba bility of a crisis.

This is because such a n environment is associated w ith great er account-

ing transparency, a large community of professional financial analysts,

respected ra ting a gencies, a w ide range of co rpora te debt securities and

derivatives demand ing sophisticated credit ana lysis, an opport unity to

make private placements, a nd efficient pro cedures for corpora te

reorganisation a nd liquidation.

In the literat ure it ha s been stressed that fa ctors w hich could

help boo st liquidity in bo nd ma rkets relate to the existence of a diversi-

fied and heterogeneous investor ba se. D iversity of investment horizons,

25 The experiences of many other Asian countries are similar to Indias;

while efforts to develop government bond markets have been fairly successful, there

has been much less success in developing corporate bond markets. Exceptions are

cases like Malaysia and Korea, where stock exchanges account for a significant

share of corporate bond trading.

Recently, a set of

comprehensive

policy measures has

been taken by the

monetary authority

for promoting the

corporate bond

segment of the

capital market,

particularly, the

secondary market

for corporate bonds.


26/27

50

Money

Finance

risk tolerance levels and investment objectives among investors pro-

vides opport unities for tra ding, w hich in turn helps enha nce market

liquidity a nd efficiency of pricing. The role of the fund management

industry is show n to be crucial in this respect, pa rticularly if regulat ors

provide them w ith incentives to tra de (M ihaljek et al, 2002 and Turner,

2002).

In India, recently, a set of comprehensive policy measures has

been ta ken by t he monetary authority f or promoting the corporat e bond

segment of t he capita l market, part icularly, the secondary m arket fo r

corpora te bond s. O ne of these is de-mat erialisation o f instruments,

w hich should go a long w ay in encouraging exchange-based trad ing of

debt securities. FIs and d ealers (bot h primary and seconda ry dealers)

ha ve been asked to make fresh investments and hold bo nds and deben-

tures, privat ely placed or o therw ise, only in dematerialised form. 26

Alongside, Na tiona l Securities Depository Limited (NSDL) and C entral

D epository Services (India) Limited (CD SL) ha ve sta rted a dmitting debt

instrument s such as debentures, irrespective of w hether these debt

instruments are listed, unlisted or privately placed. In ad dition, t he

SEBI has already ma ndat ed that a ll trades on the BSE and the NSE be

executed on the basis of the price and o rder mat ching mechanism of t he

stock exchanges as in the ca se of equities. The NSE, on the ot her hand ,

insists on credit ra ting fo r listing of all privat ely placed debt issues by

PSUs, FIs, scheduled commercial ba nks, and privat e corporat es.

While the infra structure is being developed ext ensively, in t he

corpora te debt segment, the R BIs concern, as a supervisor, remains the

large number o f priva te placements/unlisted bonds for w hich the

disclosure and d ocumentation sta nda rds are rat her unsatisfacto ry

(Reddy, 2002). The need t o have a stand ard practice in this regard

(manda tory credit ra tings), irrespective of w hether the debt is publicly

issued or privately placed, has been stressed in recent times. Such

measures should be collectively effective in ameliorating problems of

informa tion asymmetry, low liquidity a nd consequent distortions from

the corpora te debt segment, and hence help it to gro w to ma turity.

REFERENCESBhaumik, S.K., S. Bose and D . C oondoo (2003) The Emerging Indian Bond

Ma rket: Evolution, Problems and P rospects, Journal

Www.icra.in Files MoneyFinance JanMarch2003corp-Bondmarket

Documents