2

Australia
Peter McCray

I. Introduction

Australia today enjoys open, dynamic and highly e¹cient nan-

cial markets. International capital Žows are largely unrestricted, while
domestic interest rates and the exchange rate for the Australian dollar
are market-determined. Financial sector regulatory and taxation arrange-
ments have been explicitly designed so as to minimize their impact on
investment decisions and on market price signals more generally. A ro-
bust prudential framework is complemented by clearing, settlement, and
payments systems in line with best international practice.
Australia’s modern, highly developed nancial system is not a result
of accident or good fortune, but the consistent adoption of market-based
policies and a concerted program of reform that dates back at least to
the late 1970s, with the establishment of the landmark Campbell Com-
mittee of Inquiry into the Australian Financial System. The rationale
behind this longstanding policy focus on the robustness, diversity, and
Žexibility of nancial markets is not at all di¹cult to appreciate.
Among the many benets of Australia’s nancial markets is their
important role as a vehicle of real economy resiliency and macro-
economic adjustment. Accordingly, policymakers have a direct interest
in the Žexibility and capacity for innovation of domestic nancial mar-
kets, and this has attracted increasing international attention in recent
times.
For example, in an analysis presented at the Annual Meetings of
the International Monetary Fund last year, the US Federal Reserve Chairman,
Dr. Alan Greenspan, noted that those economies which had proved most
resilient in the face of the global dislocations of 1997 and 1998 were
those which enjoyed broad nancial sector diversity, and which had
available to them “multiple alternatives to transform an economy’s
savings into capital investment…should the primary form of intermedia-
tion fail.”
52 Government Bond Market Development in Asia

Dr. Greenspan went on to suggest that Australia served as “an in-

teresting test case in the most recent Asian nancial turmoil. Despite its
close trade and nancial ties to Asia, the Australian economy exhibited
few signs of contagion from contiguous economies, arguably because
Australia already had well-developed capital markets as well as a sturdy
banking system.”
This broader macro-adjustment role of nancial markets is a par-
ticularly interesting policy issue, and one on which there is doubtless a
great deal more to be said on an appropriate occasion. However, the
primary purpose of this particular paper is not to discuss Australian -
nancial markets and their broader macroeconomic signicance, but rather
to address the more specic issue of the growth and development of the
Australian government bond market.
Nonetheless, to assist in putting some of the market development
issues here into an appropriate analytical context, it is perhaps worth
sketching in a little more detail the standing of Australia’s markets gen-
erally before turning specically to a brief description of the current
shape of the domestic government bond market. These issues are cov-
ered in the following section, Part II, of the paper.
Part III looks brieŽy backward, mapping the broad institutional
terrain in which the Australian government bond market of 15–20 years
ago existed, as well as discussing some of the major operational features
of the broader nancial markets of those days, the better to illustrate the
scope and scale of reforms to institutional and market arrangements and
structures that have occurred over the intervening period. Part IV looks
in more detail at major policy and operational reforms—both in the
operational approach to markets and in supporting market infrastructure—
that have contributed to the emergence of the current market for Australian
government bonds. The paper concludes with a brief summary “check-
list” of the major market development issues that authorities might consider
in formulating strategies for establishing or building on existing domestic
government bond market capacity.

II. Australia’s Financial M ark ets: T he Current M ark et for

Australian Governm ent Bonds

Australian nancial markets are complete, in the sense that there

are observable, well-developed markets in the money and banking, bonds,
equities, foreign exchange, and derivatives sectors. None of these market
sectors is extraordinarily large judged against the size of equivalent
markets in the leading global players such as the US or Japan. The
Australian markets are not insignicant either, with trading activity in
 Australia 53

many domestic market sectors well ahead of what might be expected by

simple reference to the absolute size of the economy.
Australia has the fourteenth largest GDP in the world, amounting
to around 4 percent of global activity and across a range of nancial
products Australian markets “punch well above their weight.”
As in many other countries, the market for foreign exchange in
Australia is the largest market sector, with around half of aggregate daily
turnover in the domestic foreign exchange market involving the Austra-
lian dollar. The Australian foreign exchange market is the ninth largest
in the world, with the Australian dollar the world’s seventh most ac-
tively traded currency.
Other domestic markets are markedly smaller, and do not gure
quite so prominently on the international scale. Nonetheless, across all
sectors, Australian markets remain signicant in global terms, as the fol-
lowing table illustrates.
In the xed interest sector, the outstanding development of the
past couple of years has been the reemergence and rapid growth of issu-
ance in the domestic nongovernment bond market—at a time when the
impact of scal consolidation and the proceeds of privatization has seen
a marked reduction in the bond outstandings of both the Australian
Government and various state governments. Figures 1 and 2 highlight a
couple of interesting trends.
New domestic issues of bonds by private borrowers reached record
levels in 1999 and aggregate outstanding is now larger than those for
state government paper. Major issuers in this market are asset-backed
issuers, nancial institutions, corporates, and nonresidents. The rapid
growth of this market in recent years reŽects a variety of both cyclical
and structural factors, and prospects for further growth in this sector
seem soundly based.
However, while the reduced call on the markets by the government
sector has certainly been a factor in the rapid growth of private issu-
ance, there is also a degree of complementarity between the two market
segments—a point might well be reached where further reductions in the
supply of government bonds would not necessarily lead to even greater
issuance by the private sector. This reŽects in large part the critical role
that the government (risk-free) yield curve plays in facilitating private
issuance.
In light of this and related considerations, the Australian Govern-
ment has stated its commitment to preserving a liquid government bond
market and sovereign curve, even though there is no immediate funding
need to do so and scal projections imply the scope for early elimina-
tion of gross debt outstandings.
 T A BL E 1

54
Com parison of GDP, Foreign Turnover, Equity M ark et Turnover, and Fixed Interest Turnover in Various Countries

Gross Foreign Exchange Equity Market Turnover Fixed Interest Market

Domestic Turnover Turnover
Product a By Market By Currency Physical Market Futures Marketb Physical Marketc Futures Marketd

Government Bond Market Development in Asia

US UK US dollar US US US US
Japan US Euro Taipei, China Germany Japan UK
Germany Japan Yen Germany Japan Germany Germany
France Singapore Pound UK France Italy Japan
UK Germany Swiss franc Japan Italy France Singapore
Italy Switzerland Can dollar Switzerland UK UK Australia
China, Peoples’ Hong Kong, Hong Kong,
Rep. of China Aust dollar China Korea, Rep. of Canada
Brazil France Spain Spain Netherlands
Canada Australia France Hong Kong, Belgium
China
Spain Canada Singapore Denmark
Mexico Netherlands Switzerland Spain
Netherlands Italy Australia Switzerland
India Australia Sweden
Australia Australia
a
1998 nominal GDP; b Sharemarket index gures; c Bond market outstandings (total publicly listed); d Includes money market futures contracts.
Sources: “World Financial Markets,” JP Morgan, January 2000; Central Bank Survey of Foreign Exchange and Derivatives Market Activity,
1998, BIS; Various Exchanges, Futures and OTC World magazine; and “How Big is the World Bond Market,” Salomon Smith Barney,
July 1999.
 Australia 55

FIGUR E 1
Dom estic Bonds Outstanding

$ billion
120

100
Australian Government

80

60 State Government

40

Nongovernment
20

0
90/1 91/2 92/3 93/4 94/5 95/6 96/7 97/8 98/9

Source: Reserve Bank of Australia. Note: Turnover excludes repurchase agreements.

FIGUR E 2
Bond M ark et Turnover (daily average)

$ billion
5

Australian Government
4

2 State
Government

Nongovernment

0
90/1 91/2 92/3 93/4 94/5 95/6 96/7 97/8 98/9

Note: Turnover excludes repurchase agreements.

Source: Australian Financial Markets Association and Reserve Bank of Australia.
56 Government Bond Market Development in Asia

Although private sector bond outstandings are now approaching

the level of aggregate Australian government bond outstandings, as Fig-
ure 2 above indicates, secondary market turnover of privately issued
paper continues to be dwarfed by secondary market activity in the gov-
ernment sector.
The reduction in Australian government bond outstanding has clearly
seen some reduction in turnover in recent years, however, and some con-
solidation in the number of market makers dealing in government paper.
However, this segment of the domestic bond market remains, by a very
substantial margin, the most liquid. A couple of relevant considerations are:
First, Australia’s derivative markets in xed interest products are
deep and very liquid—the Sydney Futures Exchange, for instance, is
one of the Asia-Pacic region’s largest nancial futures and options ex-
changes. Trading in the 3- and 10-year government bond futures contracts
is particularly active.
Futures trading is the main vehicle for trading and price discovery
in the government bond market, and greatly facilitates the market-making
operations of dealers in the secondary market for such bonds. Bond
futures turnover, which has held up well in the face of declining aggre-
gate bond outstandings, is now substantially larger than turnover in the
physical market for government paper (Figure 3).
Second, consolidation of government debt into a relatively small
number of benchmark lines has also served to enhance liquidity. The
Australian government yield curve currently comprises 15 benchmark
lines across a yield curve of 12–13 years maturity. The US investment
bank, JP Morgan, calculates a liquidity ratio by looking at the propor-
tion of a government’s bonds on issue which qualify for inclusion in its
government Bond Index (as illustrated in Figure 4). The percentage of
highly liquid lines to aggregate bond outstandings in Australia remains
one of the highest in the world.
The liquidity of the market and the depth of its supporting deriva-
tive markets is reŽected in the fact that market makers are prepared to
quote spreads—between buy and sell quotes on the secondary market—
of around 2 bps on parcels of up to A$50 million of stock. This spread
is relatively narrow compared with government bond markets elsewhere
of similar size to that in Australia.
In summary, despite some reduction in outstandings and turnover
in recent years, the Australian government bond market remains highly
liquid and e¹cient, characterized by a concentrated volume of liquid
lines, a broad spread of maturities across the yield curve extending to
12–13 years, a robust secondary market and supporting derivatives frame-
work, and a broad and diversied range of investors.
 Australia 57

FIGUR E 3
Daily Average Turnover in Physical and Futures M ark ets

$ billion
7
Future Market
6
Physical Market
5

0
84/5 85/6 86/7 87/8 88/9 89/90 90/1 91/2 92/3 93/4 94/5 95/6 96/7 97/8 98/9

Note: Turnover excludes repurchase agreements.

Source: Reserve Bank of Australia and Australian Financial Markets Association.

FIGUR E 4
JP M organ L iquidity R atio (as at 31 Decem ber 1999)

Percent

100

80

60

40

20

0
Australia

Belgium

Canada

Denmark

France

Germany

Italy

Japan

Netherlands

Spain

Sweden

UK

US

Source: J.P. Morgan.

58 Government Bond Market Development in Asia

The market provides the Government with a ready capacity for

raising cost-e¬ective on-demand nancing should a future scal posi-
tion require it. In addition, the market serves as the benchmark for other
interest rates in the economy and provides a premium investment prod-
uct for a wide range of wholesale and institutional investors—banks,
funds managers and intermediaries, domestically and o¬shore. (Around
40 percent of Australian government bond outstanding is held o¬shore.)
Today’s highly developed market for government bonds di¬ers in
many fundamental ways from the market environment of 15–20 years
ago. Before turning in detail to some of the policy choices and opera-
tional decisions that have contributed to the development of the current
market environment, the next section of the paper seeks to establish
some context for those reforms by brieŽy mapping the dening charac-
teristics of government bond funding arrangements in the nancial sector
prederegulation days of the late 1970s and early 1980s.
It is important to appreciate that the substantial changes to govern-
ment bond market and funding arrangements that occurred through the
1980s took place, not in a vacuum, but very much as part of, and con-
sistent with, much broader-based reform of the Australian nancial system
generally.

III. T he Australian Financial M ark ets of 20 years ago

Up until the early 1980s, Australia’s nancial system was heavily

regulated. International capital Žows were constrained by an extensive
system of exchange controls designed to support a xed exchange rate
regime, and the domestic banking system was subject to wide-ranging
quantity controls on lending and regulated interest rates, designed to
limit the cost of nance for certain policy-preferred purposes. There were
pervasive barriers to the entry of new participants into the banking sector.
The market and institutional arrangements supporting the opera-
tion of the government bond market were very much consistent with the
philosophy underpinning this prescriptive, regulated framework.
The market was heavily oriented towards a domestic investor base,
with explicit prohibitions on most forms of nonresident entities invest-
ing in Australian government securities.
The market was also essentially “buy and hold” in its orientation,
and distinguished by a variety of “captive market” arrangements, which
obliged nancial institutions to hold specied proportions of their as-
sets in the form of government securities. Similarly, life insurance o¹ces
and pension funds were provided with signicant tax concessions in
return for holding 30 percent of their assets in public securities.
 Australia 59

One consequence of these captive market arrangements was that

there was only a very limited secondary market in government securi-
ties. Derivatives markets as they are known today did not exist.
For many years up until the early 1980s, the Australian Govern-
ment relied on a number of variants of the so-called tap method for
issuing securities. The dening characteristic of the tap issuance mecha-
nism was that the authorities determined the interest rate to apply to the
security, and investors took up the amount they wanted at that rate,
thereby determining the volume of securities that were issued.
ReŽecting the practical di¹culties of setting and maintaining the
interest rate on securities in alignment with market rates of interest, the
tap issuance mechanism gave rise to regular di¹culties in controlling
the volume of securities issued. On occasion, it saw shortfalls in funding,
when the desired amount of securities were not taken up by the market.
In such situations, the authorities used short-term borrowing from
the Central Bank to fund the shortfalls, breaching what is today re-
garded as a central tenet of government nancing—that the Government
fully fund itself in the market. It then became the Central Bank’s task to
operate in the market to o¬set the obvious inŽationary consequences of
this form of nancing, muddying the waters between monetary policy
and debt management operations.
Finally, the policy of the day was to issue a new line of stock at
each new tap, resulting in a large number of small volume lines—almost
100 separate issues in 1980. This was a far cry from today’s market
whereby virtually all government bond outstanding is concentrated in
15 highly liquid lines.

Forces for Change

The wide-ranging and complex structure of regulation that applied

right across the nancial system came under increasing stress as Austra-
lia entered the 1980s. International pressures created by deregulation,
globalization, and technological change in o¬shore markets provided a
clear impetus for reform.
As international nancial markets became more open, and interna-
tional capital Žows less restricted, the system of exchange controls was
proving increasingly ine¬ectual. This exposed the Australian dollar ex-
change rate to increasingly powerful speculative attacks and made domestic
liquidity conditions and interest rates more volatile in response to Žows
of funds across the exchange.
Domestically, it was increasingly clear that many of the direct con-
trols on the nancial system were both ine¬ectual and counterproductive.
60 Government Bond Market Development in Asia

The heavily regulated banking sector was shrinking relative to the

uncontrolled nonbank nancial sector, which Žourished, and the authori-
ties’ ability to control the growth of money and credit was undermined.
Moreover, interest rate controls on the cost of funds to “policy-
preferred” sectors, such as housing and small business, were simply serving
to limit the availability of funds for these purposes, forcing borrowers to
seek funding from higher cost alternative sources.
Financial system reform gathered pace rapidly through the course
of the early 1980s, as successive Australian governments responded to
the challenges of managing a more open economy in a more volatile
world. The dismantling of Australia’s regime of exchange controls, and
the associated Žoating of the Australian dollar in December 1983, was a
critical step and a major catalyst for further change.
Reforms to the banking sector were reŽected in two main elements:
(i) the lifting of banking controls and guidelines on deposit and loan
rates, deposit maturities, asset allocations and lending volumes; and (ii)
the relaxation of entry restrictions via the provision of banking licenses
to foreign-owned banks.
Broader-based taxation and prudential framework reforms through
the course of the 1980s contributed further to this ongoing overhaul of
the Australian nancial system. Adjustment to the new market-based en-
vironment was not without some challenges, and the late 1980s saw
some periods of turbulence as institutions adjusted to the disciplines of
the new environment, and came to terms with altered risk management
requirements as they sought to build market share within the new frame-
work.
There is no question though, from a perspective that now encom-
passes nearly 20 years of evolution and reform in the Australian nancial
system, that the exercise—which remains ongoing—has paid handsome
dividends; signicantly enhancing nancial sector competition, consumer
choice and operational e¹ciency.
Against this backdrop and, indeed, as an important part of the
wider reform exercise, a number of major changes to the conduct of
government nancing and the operation of the government bond market
were introduced through the early and mid-1980s.
The following section of the paper discusses these various changes
to government bond markets, covering the detail of the changes them-
selves, some of the considerations that led to the particular policy choices
and operational decisions that were made, and how these reforms con-
tributed to the development of today’s market for government securities.
The objective of the following section is to map the major opera-
tional and market infrastructure factors that governments may need to
 Australia 61

consider in assessing policy options for developing an e¹cient domes-

tic government bond market.

IV. T he Developm ent of Australia’s Governm ent Bond M ark et

Though it may seem an obvious point, it is worth noting that a

stable and e¹cient nancial system is an essential precondition for e¹cient
debt management and the e¹cient operation of bond markets generally.
Stability and competitiveness both lower the cost of debt nance by
reducing margins and risk premia. They also facilitate the innovation of
new products and nancing techniques, both of which increase the e¹ciency
of debt management. Development objectives in regard to domestic gov-
ernment bond markets must relate closely to the standing of the domestic
nancial system.
Many factors have contributed to the development of Australia’s
nancial system over the past 20 years. The fact that the critical markets
in Australia are based on nancing the local economy in the domestic
currency is a relevant consideration, but the adoption of market-based
policies—such as deregulation, the Žoating of the exchange rate, and
the abolition of exchange controls—has been particularly important.
Likewise, the implementation of a market based monetary policy in Aus-
tralia has provided a sound underlying monetary framework; an essential
element in the healthy development of nancial markets in any jurisdiction.
There is no question that this shift from a highly managed and
regulated, price-setting approach to funding and debt management, to
an approach fully reliant on the e¹cient workings of the market mecha-
nism has undoubtedly been the prime factor in the development of the
Australian government bond market over the past 20 years.
The very clear separation of institutional responsibility for debt
management and monetary policy functions—has also been a major con-
sideration in bringing enhanced clarity, transparency and e¹ciency to
the operation of debt management, and minimizing the potential for
markets to interpret debt management actions as carrying monetary policy
implications. Another factor has been the shift to the Government’s bud-
get funding requirements being fully funded from the market, rather than
using Central Bank nancing as a fallback.
Within this overarching market-oriented framework it is possible to
identify a wide range of policy and operational reforms introduced by
the authorities over many years, in both the operational approach to
markets and in the supporting market infrastructure, that have contrib-
uted to the development of the current highly-developed market for
government bonds. Though any attempt at categorizing these various
62 Government Bond Market Development in Asia

reforms will necessarily be somewhat arbitrary, one approach is to clas-

sify the various initiatives as being operational or infrastructure-based.

A . Operational Issues

Issuer Behavior. The main requirement of the government debt man-

ager in this respect is one of responsibility. Fundamentally, government
debt management is about managing risk. In issuing debt, governments
take on large exposures to market prices, creditworthiness and opera-
tional failure. The government debt manager’s approach to managing
these risks can send signals to the market about acceptable standards of
behavior.
The government debt manager’s behavior as a market participant is
also important to perceptions about market integrity. Investor, intermediary,
and other issuer condence can quickly be dissipated when markets are
easily manipulated and codes of conduct lax. The typical dominance of
the Government as an issuer in the domestic bond market, and its broad
regulatory responsibilities, mean that government debt management must
be conducted with the highest possible level of in-house ethics.
Responsibility also requires the Government’s debt management
activities to be transparent and reasonably predictable. However, this is
not the same as saying the Government should be completely mechani-
cal in its debt management operations. Locking debt management into a
mechanical formula may increase transparency, but may also increase
risk and cost, thereby damaging the authorities’ reputation as a debt
manager. Risk to reputation and credibility will be lowest where the
government debt manager is perceived to perform both competently and
responsibly.
In comparing the operational focus of Australian government debt
management today with that of the early to mid-1980s, two features
stand out.
First, in common with most other entities with large nancial ex-
posures in their balance sheet these days, there is a very specic focus
on the management of nancial risk in aggregate, not just on successful
execution of primary issuance. For instance, the management of the cost
and risk of the accumulated debt portfolio is just as much a priority as
funding itself.
Second, the authorities these days are much more active partici-
pants in the securities market, particularly in the area of currency and
interest rate swaps.
These features signicantly increase the onus on the authorities to
maintain regular communication with markets.
 Australia 63

With this signicantly greater direct involvement in markets, it is

critical that markets are regularly notied of, and fully comprehend,
debt management objectives and strategies, and can have complete
condence as to the motives and rationale for the authorities’ market
operation. Clearly, any taint of insider trading, or even opportunism,
would have serious implications not just for funding costs and risk pre-
mia, but also for market integrity and stability, not to mention the
Government’s reputation.
Accordingly, considerable resources are devoted to keeping nan-
cial market participants fully informed of relevant information and
developments. For a number of years now, the practice has been to make
regular public presentations to nancial market participants, setting out
details of debt management objectives, short- and longer-term strategies
to achieve those objectives, and projected borrowing programs and de-
rivatives markets activities for the period ahead. This information has
been augmented in more recent years by the production of a regular
publication—the Commonwealth Debt Management Report—which re-
ports on and reviews all operations. Regular meetings between government
debt management o¹cials and individual investors/intermediaries pro-
vide a further channel for communication.
On this issue, an interesting point of departure in Australia from
increasingly standard practice in many jurisdictions, is that the authori-
ties do not produce an auction calendar, locking the issuer into a
commitment to issuing particular stocks on particular dates in the period
ahead. Instead, the practice in Australia is only to commit to conduct
bond auctions within a broadly dened issuance window, currently ev-
ery four to six weeks. This greater Žexibility still provides some measure
of certainty to the market as to the timing of prospective supply, but
without exposing the issuer to excessive event risk should a particular
auction date prove unattractive in light of market developments. In ad-
dition, the detail of the stocks to be issued is announced only on the
day before the auction.
This is a useful practical illustration of the need for the authori-
ties, in managing relations with market participants, to strike an appropriate
balance between a desirable and mutually benecial level of transpar-
ency as to issuance intentions on the one hand and, on the other, using
an excessively mechanical and formula-driven approach to the point of
nancial detriment.
The very rigid approach to auction schedules favored by govern-
ment issuers in some of the larger sovereign bond markets is not felt to
be appropriate to Australia, where greater Žexibility in determining the
precise timing and make-up of auctions is preferred. There is no evidence
64 Government Bond Market Development in Asia

at all that any sort of risk premium attaches to bond yields in Australia
as a result of this practice.
This is a good illustration of the important point that sovereign
debt management, and more particularly the development of sovereign
bond markets, is not merely a matter of mimicking the strategies and
conventions applied in the largest and most liquid bond markets.
Relevant considerations that need to be carefully assessed in for-
mulating development strategies are the standing of the market in global
terms, the quality of the existing market infrastructure, the characteris-
tics of the investor base, the scal impact of debt-service costs, and
market perceptions regarding the sustainability of the debt burden.
Many market development issues are essentially practical ones, and
it is simply not sensible to believe that a “one size ts all” blueprint
can successfully be applied to all sovereign debt managers, in all cir-
cumstances. There will always be a need for sound judgment in determining
precise operational strategies and conventions and, where necessary, the
balance to be struck between competing considerations to best meet the
requirements of individual markets.
What is certain, however, is that an ethical, responsible approach
to sovereign debt management, and a culture of transparency and com-
munication with markets, is an essential prerequisite to market development.

B. Prim ary Issuance M echanism

For many years up until the early 1980s, the authorities in Austra-
lia used a tap mechanism as the primary vehicle for issuing debt securities.
They determined the price (interest rate) at which the securities would
be o¬ered to the market, and the market took up as much of the stock as
it wished at that price.
The tap system operated reasonably successfully through the 1950s
and 1960s when nancial markets were regulated. There was little compe-
tition from private sector borrowers and good support from captive investors.
The system proved increasingly unsatisfactory through the 1970s, how-
ever, with uncertainty surrounding fund-raising the major drawback.
By the early 1980s, it was clear that an alternative mechanism was
required. In principle at least, there were two broad choices open to the
authorities.
The rst was an outright move to issuing bonds by auction, under
which the market clearing price for the volume of bonds made available
for issue would be determined by the bids of market participants.
The second option was the use of a “dealer panel” mechanism,
whereby appointed members of a panel would be obliged to take up
 Australia 65

new stock at an agreed price (and, typically, to make markets-quote

continuous two-way prices—in the secondary market as well).
In practice, there are many possible forms of dealer support that
fall under this broad heading, ranging from fully underwritten arrange-
ments to less formal “bought deal” arrangements, whereby each member
of the panel agrees to buy a certain quantity of stock at a certain price
before seeking to place it in the market on a best endeavors basis.
These dealer panel arrangements can o¬er some advantage in certain
circumstances, for instance where there is a concern that low initial li-
quidity might act as a disincentive to bid in an auction. However, cost
(i.e., fees) can be considerable, particularly for a fully underwritten issue,
making the mechanism largely impractical for an ongoing issuance program.
In 1982, the Australian authorities moved to issuing bonds by auc-
tion. This was, and remains, the technique favored by many OECD
governments for issuing large volumes of securities into reasonably well-
established markets. The judgment in Australia by the early 1980s was
that the government bond market was su¹ciently liquid, and the mar-
ket-making community and secondary market su¹ciently well-established,
for the auction method to be successful.
Initial experience with the move to the auction issuance mecha-
nism was certainly a little mixed, with a wide range of bids accepted in
a number of the earliest tenders, as Figure 5 illustrates.
Such results were not unexpected, however. The feeling was that
these early mixed results were a “price worth paying” for the broader
benets to come in time from a move to a fully market-based auction
mechanism. The judgment proved well-founded, with auction results
improving steadily as the market deepened. The use of auctions became
widespread internationally during the 1980s, as nancial markets be-
came less regulated and more sophisticated, and more dominated by
large professional institutional investors.
The move to bond auctions in Australia removed the uncertainty
surrounding fund raising, and was e¬ective in enabling the authorities
to sell large quantities of stock with minimal disruption to interest rates.
Secondary market activity in bonds increased further, encouraged by the
adoption of the auction mechanism and by the growing volume of stock
on issue. With the removal of restrictions on nonresident investment in
Australian government bonds, o¬shore interest in the market increased
signicantly. The adoption of an issuance mechanism in line with major
OECD bond market norms was no doubt an encouraging factor.
The use of auctions for Australian government bonds is now, of
course, very well-established, and there are no plans to consider alternative
issuance approaches.
66 Government Bond Market Development in Asia

FIGUR E 5
Bond Auctions: R ange of A ccepted Bids

basis points
140

120

100

80

60

40

20

0
1 9 17 25 33 41 49 57 65 73 81 89 97 105 113 121 129 137 145 153 161
Tender Number
1982 1987 1992 1995 1997 1999

Source: Reserve Bank of Australia.

However, it is not true to say that the auction method is to be pre-

ferred in all circumstances. The most e¬ective primary issuance method
depends on a wide range of considerations, such as the predominant
type of investor (for example large institutional or “retail” individual)
and the stage of development of the bond market.
Certainly, the dealer panel mechanism is not without merit in cer-
tain circumstances, and may be particularly useful when issuing a new
line of securities, where investor demand may initially be low, as it
ensures the success of the issue. It can also help to introduce new debt
instruments where lack of investor familiarity with the product and/or
concerns regarding the lack of an established secondary market again
leave the authorities open to the risk of weak bidding in auctions.
The Australian authorities used a form of dealer panel mechanism
for a couple of years in the early 1990s when launching an indexed
bond issue program, and this approach was successful in the early build-
ing of an investor base and a number of market makers in this form of
instrument.
However, a move to an auction mechanism for primary issuance of
indexed securities followed once it was felt that the market had devel-
oped su¹ciently in liquidity and depth to successfully support such an
approach.
 Australia 67

Basically, the selection of primary issuance mechanisms must take

into account the stage of development of the market. Open auction ap-
proaches are particularly appropriate in reasonably well-established, liquid
markets with established secondary markets and market makers. Smaller
or nascent, less liquid markets may benet from closed dealer panel
arrangements, at least initially and while issue volumes are not so large
as to make the cost of such panel arrangements prohibitive. It will likely
be appropriate to move to auction arrangements as the market develops
and matures, however.

C. Captive Investor A rrangem ents

As noted earlier, captive investor arrangements were a pervasive

feature of the Australian government bond market up until the early to
mid-1980s.
The aim of these arrangements was not only to create demand for
government bonds. Captive investor arrangements were also motivated,
at least partly, by prudential concerns. There is clearly a prudential ben-
et in forcing institutions to hold minimum levels of the highest-credit
securities. However, these arrangements also ensured a continued de-
mand from growing nancial institutions for government securities, and
doubtless led to the authorities issuing government bonds at lower inter-
est rates than would otherwise have been the case.
In principle, therefore, captive investor arrangements might be con-
sidered to have some value in building government bond market capacity.
Where bond markets are not highly developed and/or there is no high
demand for government paper, such arrangements might be considered
as acting as an e¬ective, if crude, discipline on markets. However, cap-
tive arrangements can also be very costly and ine¹cient, from a number
of perspectives.
Such arrangements provide governments with the scope to raise
funds comparatively cheaply, removing an important scal discipline,
and encouraging governments to be less careful in their spending decisions.
Captive investor arrangements serve to distort the Žow of invest-
ment funds in the economy and act as a tax on the captives, by reducing
their capacity to earn higher returns on alternative investments. Perhaps
most importantly, such arrangements inhibit the development of second-
ary markets in the relevant securities. In a deep and liquid secondary
market, investors and/or traders must be able to take both short and long
positions in securities.
Clearly, under captive holder arrangements, the captives are unable
to take short positions, and the fact that these arrangements tend to hold
68 Government Bond Market Development in Asia

the yields on relevant securities articially low means that the holding
costs of taking long positions can be high. Captive investor arrange-
ments therefore discourage the taking of positions in the market and act
to inhibit liquidity and secondary market development.
In common with the experience of all major OECD jurisdictions,
there are no longer any captive investor arrangements applying to the
holdings of government bonds in the Australian market. Experience makes
it fairly clear that there are likely to be substantial net costs in employ-
ing captive investor arrangements as part of a government bond market
development strategy.

D. Prom otion of Secondary M ark et and Derivative M ark ets

A vibrant secondary market for government bonds o¬ers a number

of well-known advantages. The e¬ectiveness of a deep and liquid sec-
ondary market in increasing demand for government securities and, thus,
reducing a government’s cost of funds, provides ample incentive for the
authorities to take an active interest in building secondary market ca-
pacity.
The secondary market for Australian government bonds is large
and very active, and has expanded substantially since the prederegulation
days of the early 1980s. Government bond turnover data provide a good
indication of trends in secondary market activity over this period.
The graph highlights the point that the single most potent inŽu-
ence on the level of activity in secondary markets is the volume of new
issuance and bond outstandings. The past few years have been the only
period since the early 1980s where Australian government bond market
turnover has declined substantially—a period in which the Government
has run sustained scal surpluses and the stock of bonds outstanding
has consequently declined.
Bond market turnover currently remains somewhat higher than the
levels of the mid-1990s, however, even though the stock of gross bond
outstandings currently has hardly changed.
The implication is that, beyond simply issuing large volumes of
debt into the market—ultimately, a development tool beyond the con-
trol of the debt management authorities—there is much else that the
authorities can do to stimulate the development of a secondary market.
The quality of overall market infrastructure—for example, the
e¹ciency of clearing and settlement arrangements, and the robustness of
the prudential/regulatory regime—is clearly critical. A number of factors
already touched on also have a bearing on secondary market develop-
ment, including market perceptions as to the credibility and reliability
 Australia 69

FIGUR E 6
Daily Average Turnover versus Bonds Outstanding

$ billion $ billion
100 7
Bonds Outstanding (LHS)
6
80 Turnover (RHS)
5

60 4

40 3

2
20
1

0 0
84/5 85/6 86/7 87/8 88/9 89/90 90/1 91/2 92/3 93/4 94/5 95/6 96/7 97/8 98/9

Note: Turnover excludes repurchase agreements.

Source: Reserve Bank of Australia and Australian Financial Markets Association.

of the issuer and issuance programs, the form of primary issuance ar-
rangements adopted, and the prevalence of captive investor arrangements.
Beyond these generic factors, there are a number of additional spe-
cic considerations discussed below that may also be of relevance to
building secondary market capacity in government bonds.
Certainly, it is most important to issue securities with the broad
characteristics that the market demands. Close contact between investors
and market makers is required, to monitor the popularity of the various
stocks on issue and, if necessary, consider buying back lines of stock
that are no longer in demand and replacing them with more popular
securities.
The liquidity of individual lines is a critical consideration for in-
vestors. In Australia, a large-scale bond consolidation program in the
late 1980s saw the number of individual bond maturities reduced from
almost 100 comparatively illiquid lines, to the point where today almost
all bond outstandings are concentrated in 15 highly liquid benchmarks.
Outright liquidity is not the only consideration, however. Careful
selection of the maturity of lines to be issued along the sovereign curve
and, generally speaking, the maintenance of bond coupons reasonably
close to prevailing market yields, are other important considerations for
many investors.
70 Government Bond Market Development in Asia

The encouragement of specialist market makers in government bonds

provides a further vehicle for promoting secondary market activity. There
need not be an excessively formal arrangement with the authorities, though
arrangements in many sovereign jurisdictions do tend to be quite structured,
with primary dealers (PDs) obliged to bid in auctions as well as make
continuous two-way prices in the securities of the relevant authority. In
many instances, the authorities dictate the spread between bid and o¬er
prices that market makers are to quote. Fees, or some form of “preferred
access,” can often be required for the provision of such services.
Although the Australian practice will not necessarily be applicable
in other government bond markets, the experience here is that highly-
structured, fee-based arrangements are not essential in establishing a
successful government bond market-making infrastructure. Be that as it
may, the present arrangements in Australia do reŽect the outcome of a
deliberate strategy instigated in the early 1980s to build secondary mar-
ket capacity in government bonds. An integral element of that strategy
was the formation of a specialist Reporting Bond Dealers group, with a
select market membership, which met regularly with the authorities to
exchange market information and discuss current developments.
Eligibility for membership of the group was determined solely on
the basis of the volume of trading in government bonds undertaken, and
was reviewed annually. Aside from this annual review process, there were
no prudential or other requirements placed on members of the group.
Although the Reporting Bond Dealers enjoyed no particular privi-
leges or special relationship with the Government beyond the functions
noted, the prestige and status associated with membership of the group
served to add a useful further element of competition to the market. The
Reporting Bond Dealer arrangements were dissolved in 1992, by which
time the infrastructure for market-making Australian government bonds
had been long and successfully established in the market. Under current
arrangements, by convention, a pool of self-selected market intermediar-
ies (currently numbering around 10) agree among themselves to make
continuous two-way prices in government bonds to each other—and, by
and large, to clients—at a 2 bp spread between bid and o¬er yields.
The authorities themselves can, where appropriate, play a more
direct role in encouraging secondary market development through the
provision of stock lending or repurchase agreement (repo) facilities in
government bonds. Of course, market manipulation motivated by opportunism,
or the pursuit of short-term prot-making, is completely unacceptable in
these circumstances.
And even where some form of intervention is fully warranted by
prevailing circumstances, considerable care will be required on the part
 Australia 71

of the authorities in directly involving themselves in the secondary mar-

ket for their own securities. As discussed already, clear communication
with the markets as to objectives, strategies and motives is essential.
Finally, where an e¹cient and vibrant intermediary market-making commu-
nity is well-established and the Government is a steady issuer into the
primary issue market, o¹cial sector involvement will likely be relatively
marginal in terms of bond market activity overall.
However, the provision by the authorities in the appropriate cir-
cumstances of stock lending and/or repo facilities to the market can
encourage both more e¹cient market making by intermediaries and greater
market participation by investors, who will draw comfort from the capac-
ity of the authorities to manage occasional liquidity squeezes or other
periodic market imperfections.
Further, the Australian experience is that allowing the short selling
of government bonds (in appropriately prescribed circumstances) can also
contribute substantially to building trading activity and market liquid-
ity. (Short-selling occurs when an entity sells securities to a purchaser
without, at the time of the sale, having the right to transfer ownership of
the securities to the purchaser.) If short-selling is not permitted, market
makers can make two-way prices only in securities which they own,
which clearly reduces competition and liquidity.
The scope for short-selling, and thus building market activity and
liquidity, is greatly enhanced where there are established markets for
stock lending and repo transactions. In particular, repo agreements in-
crease the e¬ective amount of securities available for trading by enabling
securities to be traded without longer-term holders having to give up
ownership beyond the period of the deal.
It is worth emphasizing that a framework which allows for short
selling of government bonds must be backed by a rigorous prudential/
regulatory regime that comprehends the scope for abuse of such arrange-
ments, and makes very clear the precise circumstances, and under what
conditions, such short-selling is permitted.
Finally, the presence of liquid and e¹cient repo and derivatives
markets—markets in bond repo agreements and bond futures, in for-
wards, in swaps and other forms of derivative of the underlying government
bond product—can contribute signicantly to building secondary mar-
ket activity in the physical government bond.
Repo markets both underpin market making by intermediaries
through, for instance, facilitating short-selling of securities, and also in-
crease the attractiveness of the underlying physical bond in the hands of
investors by providing them with the scope to “gross up” their holding
return while retaining ownership of the security longer term.
72 Government Bond Market Development in Asia

E¹cient derivatives markets provide a vehicle for investors in gov-

ernment bonds to manage their interest rate risk—the risk that interest
rates in the future will be di¬erent from today’s. Derivatives markets
allow market participants to x today the prices at which trades will be
made in the future to hedge their underlying exposure to future price
movements. Accordingly, the availability of these markets for risk hedg-
ing promotes increased market turnover. Intermediaries and investors are
more willing to buy, sell and trade, and bid/o¬er spreads are tighter in
the market for the underlying physical bond.
In Australia, there are highly-developed and active markets in both
repos and a wide range of derivatives products. The general philosophi-
cal approach to developing these markets has been one of facilitation—of
providing the kind of legislative and regulatory environment that gives
market participants the condence to put capital at risk, and the kind of
market-based framework that encourages markets to develop at a pace
and in a direction fully consistent with commercial considerations.
Facilitation can, however, take more immediate forms as well. The
development of the 3- and 10-year bond futures contracts on the Sydney
Futures Exchange has been supported by the authorities through careful
consideration of futures basket requirements when determining primary
bond issuance and repurchase strategies. For example, before issuing a
new benchmark line, careful consideration is given to the underlying
bond basket requirements of the relevant futures contract.

E. Infrastructure Issues

Clear Separation of Debt M anagem ent and M onetary Policy R espon-

sibilities. As already mentioned, the separation between these two
functions in Australia is now complete. Not only does the Government
fully meet its debt nancing requirements from the market, and at mar-
ket determined rates, but there is a clear institutional separation as well,
with monetary policy and liquidity management the responsibility of
the Reserve Bank and debt management the responsibility of the Austra-
lian O¹ce of Financial Management.
This has not always been the case, however. Indeed, the combina-
tion of a managed exchange rate regime and the tap mechanism for the
issuance of government bonds that existed in Australia up until the early
1980s resulted in a very tight link between monetary policy and debt
management.
Under the xed exchange rate regime, the need for the Reserve
Bank to stand in the market for foreign exchange, ready to buy or sell
foreign currency at a xed price against Australian dollars, meant that
 Australia 73

foreign currency debt and Australian currency debt were regarded as

very close substitutes by the market. Thus, attempts by the Reserve Bank
to tighten monetary conditions by selling holdings of government debt
from its own portfolio into the market were readily o¬set by reduced
holdings of foreign debt, with little net impact on base money.
Likewise, under the tap issuance mechanism the yields set by the
Government to apply to new debt issuance became critical for monetary
management. If tap yields on new bond issue were held constant, but
the Reserve Bank wished to tighten monetary policy, the objective of
the policy adjustment would soon be thwarted.
While the tightening of policy would lead to a rise in short-term
private interest rates, this would serve only to make holding government
bonds a less attractive proposition. The Government’s budget funding
shortfall would therefore widen until the Bank was called upon to meet
the funding shortfall. This nancing of the government funding need
would increase the supply of liquidity in the market, making the entire
process self-defeating.
The need to o¬set the domestic monetary e¬ects of foreign exchange
inŽows or outŽows provided yet a further complication. On occasion,
where these Žows were particularly large, the authorities would be obliged
to use primary issues of government bonds not only to nance the budget
(debt management), but also to try and control the volume of liquidity
in the domestic money market (liquidity management).
The e¬ect of this untidy framework was that bond issuance under
a xed exchange rate regime and the tap issuance mechanism became as
much an arm of monetary policy as of debt management. Indeed, it was
not uncommon for announcements of an increase in interest rates on
new debt issues to be interpreted as adjustments in the stance of mon-
etary policy.
Of course, in Australia, the longstanding move to a strict opera-
tional and institutional separation of responsibility for debt management
has long since addressed these gross operational ine¹ciencies. But such
ine¹ciencies—serious though they are—account for only part of the
rationale for a segregation of policy responsibility.
Other cogent reasons include the reduced scal discipline associ-
ated with a government’s capacity to raise cheap funds from the Central
Bank, the likely inŽationary consequences of this form of “o¹cial sec-
tor” funding, the adverse implications for the development of a viable
private secondary market in government bonds, and the overall loss of
clarity and predictability in debt management operations. It is with good
reason that sound nancial management is now widely accepted to re-
quire the two activities to be kept separate.
74 Government Bond Market Development in Asia

F. Clearing and Settlem ent System s

Reliable and timely clearing, transfer of ownership, and settlement

arrangements are clearly essential to the e¹cient and e¬ective operation
of securities markets. A rigorous and reliable clearing and settlement
infrastructure allows market participants to undertake bond market trans-
actions without undue risk—whether that be default risk, market risk or,
indeed, systemic or other broader risks. Accordingly, the e¬ectiveness of
such systems can be a signicant inŽuence on the development of sec-
ondary market activity.
In Australia, a gradual move from the mid-1980s towards an elec-
tronic infrastructure for clearing and settlement arrangements, building
on an already robust base, has contributed in no small way to the devel-
opment of the government bond market.
The Reserve Bank has operated an electronic transfer and settle-
ment system for Australian government securities—known as the Reserve
Bank Information and Transfer System (RITS)—since 1991. RITS, which
among other considerations introduced the delivery versus payment (DvP)
framework for the settlement of government bond transactions, repre-
sented a considerable advance on previous manual systems. It represents
both greater e¹ciency and enhanced risk management for all market
participants.
The RITS platform enables the full range of transactions in Austra-
lian government securities, including delivery of bond auction proceeds
and secondary market trades in government bonds, to be settled remotely
via terminals in members’ back o¹ces. Virtually all signicant market
participants are members of RITS, and current membership stands at about
240. Remote settlement can, in turn, lead to signicant cost saving prac-
tices, such as “straight through processing.”
More recently, RITS and other market electronic settlement sys-
tems have been interlinked under the aegis of the introduction of the
Real Time Gross Settlement (RTGS) system into the Australian markets,
yielding further e¹ciencies and risk management enhancements. Australia’s
RTGS system is among the most advanced in the world because of the
way it closely links to a variety of securities settlements systems, giving
it a very high coverage.
Very broadly, the essential features of a robust clearing and settle-
ment infrastructure can be summed up as follows: (i) clear and unambiguous
regulations of systems setting out, for example, eligible instruments,
detailed operational procedures and when a transaction is nal; (ii) sound
legal underpinning of regulations—providing legal certainty in all rel-
evant jurisdictions, for example in regard to the enforceability of default
 Australia 75

procedures; (iii) sound risk management procedures in the system to

limit the risks of loss of cash or securities paid to a defaulting counterparty
prior to the detection of the default, for example by introducing DvP;
(iv) well-established and understood contingency arrangements and fallback
procedures to deal with circumstances of operational or technical fail-
ure; and (v) explicit identication of the government institution responsible
for regulatory oversight (for example, in relevant legislation), which should
be appropriately empowered to obtain relevant information and give
enforceable directions.

G. R egulatory Fram ework

A well-functioning bond market requires a regulatory system which

provides a level playing eld, clearly dened property rights, a transpar-
ent information Žow, and a capable regulatory authority. In Australia,
the Corporations Law includes provisions relating to the trading of gov-
ernment and corporate debt. The majority of provisions relating to
obligations of market participants apply equally to the government and
corporate bond markets. For example, prohibitions against engaging in
misleading and deceptive conduct, stock market manipulation, false or
misleading statements, and a requirement for dealers to be licensed are
imposed equally on participants in both markets.
The Corporations Law also governs the fund-raising of corporate
issuers. This ensures, through the imposition of disclosure and prospec-
tus requirements, that market participants purchasing such securities are
able to make informed investment decisions. Although not formally cov-
ered under Corporations Law provisions in these areas, the Commonwealth
has adopted “best practice,” so that its prospectuses clearly inform po-
tential investors of the terms and conditions of issuance and the
Commonwealth’s obligations to the investor for the payment of interest
and the repayment of principal.
The legislative framework is, in turn, superimposed on a frame-
work of self-regulation, coordinated by bodies such as the Australian
Financial Markets Association, and representing the OTC nancial mar-
kets on matters such as market codes of conduct and participant standards
and accreditation.
The regulatory regime cannot of course remain static, but must
keep pace with changes in the nancial service industry—in customer
needs, competition and product development and distribution. In this
regard, the Government is developing legislation to establish an inte-
grated regulatory framework for Australia’s nancial products markets,
clearing and settlement facilities, and nancial service providers.
76 Government Bond Market Development in Asia

V. Conclusion

This Paper has addressed a range of issues surrounding the devel-

opment of the Australian government bond market over the past 20 years.
A particular focus has been to identify the major reforms—both in the
operational approach to markets and in supporting market infrastructure—
that have contributed to the emergence of the current highly-developed
market for Australian government bonds.
This review of Australia’s experience suggests a number of conclu-
sions of possible relevance to other jurisdictions seeking to formulate
strategies for establishing or building on existing domestic government
bond market capacity.
Issuers must act responsibly at all times, and with the highest pos-
sible level of in-house ethics. Regular and substantive communication
and dialogue with markets as to debt management objectives and opera-
tional strategies is paramount. The rationale for debt management operations
should be transparent and the operations themselves reasonably predict-
able, but not so completely mechanical that the Government is exposed
to unacceptable risks in debt management performance.
Selection of primary issue arrangements should take into account
the stage of development of the market. Generally, the open auction
mechanism is preferable, though smaller, less liquid markets may benet
from some form of dealer panel arrangement in some circumstances, at
least until the market develops and matures—and so long as issuance
volumes are not so large as to make cost prohibitive.
Captive investor arrangements are not recommended. There are likely
to be substantial net costs in employing such arrangements as part of a
government bond market development strategy.
Markets require a steady supply of new securities to sustain liquid-
ity. Secondary market liquidity can be encouraged by active management
and support by the authorities. This support might include provision of
stock lending facilities, the consolidation of bonds into highly liquid
lines, management of market-making arrangements and the establishment
of a market environment conducive to the successful operation of repo
and derivatives markets.
A clear separation of debt management and monetary policy re-
sponsibilities, and full funding of government budgets from the market—at
market determined prices—is essential, as are reliable and timely clear-
ing and settlement arrangements.
A well-functioning government bond market requires a prudential/
regulatory regime that provides both legal certainty and a level playing
eld, and remains alive to the changing requirements of a dynamic market
environment.
 Australia 77

Finally, the ultimate threshold issue in seeking to make participa-

tion in the government bond market an attractive proposition for investors
and intermediaries is creating an attractive economic case for doing so.
No matter how robust a market infrastructure, or how e¹cient its opera-
tional features might be, these considerations will not of themselves
generate market development momentum if the product itself is unat-
tractive in economic terms.