Speech Recent Developments in Asian Bond Markets
Assistant Governor (Financial Markets)
Talk to the 17th Australasian Finance &
Following the Asian crisis, there was widespread agreement that Asian economies would benefit from having larger and more developed bond markets. In response, authorities in the region began work, both at the national level and the regional level, on a wide range of initiatives in pursuit of that objective.
Many of these initiatives are still in their early stages, but already there has been substantial development in regional bond markets in terms of both size and depth. In several respects, however, regional markets continue to lag behind those in some of the mature economies, leading commentators to question some of the initiatives on the grounds that they have been ineffective or not well-sequenced.
This criticism is, I feel, partly based on unrealistic expectations. It may also reflect a misunderstanding of what the authorities are trying to achieve. I thought it might be useful, therefore, to run through recent developments in Asian bond markets, outline the various initiatives that are under way, and try to address some of the criticisms that have been made.
Bond Financing in Asia
At the time of the crisis, Asian economies on the whole were very dependent on their banking systems for the provision of finance. Banks provided about 80 per cent of the debt financing in the economy, while bond markets provided only 20 per cent. (Graph 1)
The proportion of bond financing was low not only by the standards of mature economies, but also relative to other emerging markets. In mature economies, bond financing typically accounts for about 30–40 per cent of total debt financing, though as much as 60 per cent in the US.
That the Asian region fell short of the figures in the mature economies should be of no surprise, since the share of bond financing tends to increase as economies mature. There are various reasons why this is so. For one thing, bond markets require an extensive infrastructure, including well-developed accounting, legal and regulatory systems, payments and settlements systems, ratings agencies, networks of brokers to sell bonds and so on. Bond financing is also more suited to well-established companies whose operations and credit standing are well known. In emerging markets, where the corporate sector is developing rapidly and small and medium enterprises dominate, it is understandable that banks have a comparative advantage in the provision of financing, as they are better able to know the borrower's business and hence assess the credit risk.
The fact that the proportion of bond financing in Asia was lower than in other emerging markets appeared, on the surface, to be more noteworthy, though an important part of the difference was due to the fact that Asian governments had much less debt on issue than governments in other emerging markets.
The points I have just made describe the situation in non-Japan Asia as a whole, but there was in fact quite a variation among countries in the Asian region. South Korea, Malaysia and the Philippines, for example, had quite high bond financing ratios even in the pre-crisis period. (Graph 2)
The onset of the crisis in 1997 quickly exposed the risks that an economy faces by having the bulk of its financing eggs in the banking basket. If something goes wrong in the banking sector, the overall flow of financing to the economy becomes restricted as there are no alternative financing avenues. The problem is made particularly acute by the fact that bank finance can be relatively short term, so not only is the flow of new loans reduced, but existing loans can also be cut back quickly.
Graph 3 shows how quickly and sharply growth in bank lending slowed following the crisis. Many Asian economies experienced falls for a number of years in the amount of bank loans outstanding, and in some cases growth in bank credit has yet to recover fully.
Bank lending is particularly vulnerable to cutbacks where foreign banks are involved, since international bank lending appears to be especially subject to fads. In the case of Asia, for example, foreign banks committed vast amounts of money to the region in the lead up to the crisis – more than the region could usefully employ – but withdrew it quickly once the crisis hit. (Graph 4) Such large fluctuations in the availability of finance are, of course, highly disruptive to the economy.
It is not surprising, in view of the events that unfolded, that there emerged strong support throughout the region for the development of financing through alternative channels to the banking system. This policy direction was also strongly supported by the international institutions such as the IMF and the World Bank, as well as a range of analysts and commentators around the world.
Bond Market Developments since the Crisis
Asian bond markets have come a long way since the crisis. I should clarify at this point that I am focussing on the market for local currency bonds issued by governments or corporations as opposed to bonds issued in foreign currencies. The amount of such bonds on issue has more than doubled in size since 1997. The increase has been particularly apparent for government bonds, as governments in many Asian countries have run significant budget deficits since the crisis, but issues of private sector bonds have also increased notably. Total local currency bonds on issue in non-Japan Asia currently stand at about US$1,400 billion, or about 50 per cent of the relevant countries' GDP. (Graph 5) Growth in bond markets has been widespread across the region. (Graph 6)
As a result of this growth in the region's local currency debt markets, they now account for around 3 per cent of such markets worldwide, double the share of Latin America and almost six times as large as that of Eastern Europe. (Graph 7) While impressive, this figure needs to be qualified to some extent as it overstates the universe of Asian bonds available to investors because it includes some illiquid bonds issued for bank recapitalisation purposes and held in special purpose vehicles. Also access to bonds by foreigners, particularly in China, though in some other countries also, is limited by exchange controls.
Despite the growth that has taken place, Asian bond markets are still largely over-looked by international investors. Foreigners own less than 5 per cent of Asian bonds on issue, quite low by the standards of many other countries. The corresponding figures for Australia and the US, for example, are roughly 25 per cent.
Part of the problem may well be that Asian bond markets are still perceived to be relatively illiquid. Annual turnover of bonds is quite low, typically in the order of one or two times the amount of bonds on issue, though Hong Kong, Singapore and Korea have turnover ratios that are closer to those in mature economies. (Graph 8)
Various factors may be contributing to this lack of liquidity. One is that a large proportion of bonds is held by banks, often under regulatory arrangements. Across the region, banks hold over half of government bonds on issue, which contrasts with a figure of less than 5 per cent in the US and Australia. In some Asian countries, the central provident fund also tends to be a large holder of bonds, particularly government bonds. These types of institutions tend to ‘buy and hold’ and therefore do not promote market turnover. Compounding the problem is that holders may not be required to mark their bonds to market, which substantially reduces the incentive to turn the bonds over.
Even though Asian bond markets have developed rapidly since the late 1990s, it is widely recognised within the region that there are still shortcomings and more work needs to be done. The authorities are pursuing this both at the national and regional levels. A high priority for many national authorities has been to put in place arrangements for the issue of government debt which reflect best practice from around the world. For example:
- all major countries in the region have moved to tender based systems for issuing government debt and have established a yield curve, in some cases out to 20 years;
- many now have pre-announced issuance calendars, and tend to issue into a few benchmark lines, with various bonds being reopened in successive tenders in order to build depth in individual issues;
- five countries have established futures markets for government debt; and
- three countries have repo markets.
At the regional level there are three main government-sponsored initiatives underway, aimed at promoting the development of local currency bond markets. The benefit of this regional approach is that it allows individual economies to leverage off each other in implementing reforms, creating momentum. It also contributes towards harmonising regulatory and legal structures and market practices.
One set of initiatives is under the APEC banner. The main focus of these is to promote securitisation and credit guarantee markets. A particular issue this work is trying to address is the fact that the credit standing of borrowers in the region is often lower than the requirements of investors, especially some institutional investors.
A second initiative is being progressed through ASEAN+3. This work is also focused on identifying impediments to market development, but across a broader range of issues than the APEC initiative. Six working groups, each with a focus on a specific aspect of the market (e.g. credit rating agencies, settlement procedures and credit guarantee arrangements) have been established to develop strategies to remove impediments.
The third initiative is the work being undertaken by EMEAP. EMEAP is a grouping of eleven central banks in East Asia and the Pacific, comprising Australia, China, Hong Kong SAR, Indonesia, Japan, Korea, Malaysia, New Zealand, the Philippines, Singapore and Thailand.
This initiative differs from others in that it has sought to take a practical approach, through direct participation in markets, to identifying obstacles to bond market development and to building an infrastructure capable of supporting bond markets. Specifically, the central banks felt that a good way to expose any problems to market development would be to work together to set up, and invest in, a bond fund. It was also felt that some of the work in setting up such a fund would help to create an infrastructure which may have a more general application in bond markets.
The first phase of this initiative involved the setting up of the Asian Bond Fund, known now as ABF1. This was launched in the middle of 2003.
ABF1 is a fund which pooled US$1 billion in US dollar reserves from the EMEAP member central banks for investment in US dollar-denominated bonds issued by eight of EMEAP's governments – China, Hong Kong, Indonesia, Korea, Malaysia, the Philippines, Singapore, and Thailand. The fund does not include bonds issued by Australia, Japan and New Zealand, as these economies already have relatively deep and broad markets.
ABF1 has been operating successfully now for about 18 months. It was never seen as an end in itself, since its potential contribution to market development was always going to be limited given that it was confined to US dollar securities and that the participating investors were limited to the eleven central banks. Rather, it was seen as a trial run for a more complex fund based on local currency bonds and involving a wider range of investors, including central banks from outside the region as well as private sector investors.
Work on this second fund, known as ABF2, commenced soon after the completion of ABF1 was announced. Because of its more complex and technically demanding nature, it has required a larger commitment of resources by the participating central banks than was the case for ABF1. Details of the fund are to be announced soon, but already it can claim success at one very important level, in that it has highlighted how effectively central banks in the region can work together to achieve a common goal.
The framework for ABF2 consists of two components:
- a single Pan-Asian Bond Index Fund (PAIF); and
- a Fund of Bond Funds (FoBF) which consists of investments in eight separate country sub funds.
Each of these funds will be managed by private sector funds managers and benchmarked against a new Asian government bond index.
Unlike ABF1, which is confined to central banks, investment in ABF2 will in due course be open to private sector investors, providing them with low-cost vehicles to invest in Asian bond markets of their choice. The PAIF will cater to investors who wish to have a well-diversified exposure to bond markets across the eight East Asian markets in a single investment, while the individual country sub funds will allow investors to tailor the composition of their investments across the region.
The work that is being undertaken on bond markets in the region has attracted a lot of international attention. This is overwhelmingly a good thing as it has resulted in a lively interchange of ideas as well as the compilation of a large amount of data on regional markets. Anyone conducting an internet search on Asian bond markets will now be confronted with over 500,000 papers and notes on the topic. Among the official sector, comprehensive studies of Asian bond markets have been published by the Asian Development Bank and the Bank for International Settlements. We at the Reserve Bank of Australia have also contributed with an article in the December 2003 edition of our Bulletin.
Not all of the commentary that one hears about Asian bond markets has been positive. Criticisms are still made, for example, about the small size and lack of liquidity of the markets. Often, however, the basis of such criticisms is comparisons with highly developed markets elsewhere, particularly the US, and as such is probably unfair. It is unrealistic to expect that Asian markets would measure up well against the US, since few markets do, even among the mature economies.
It also needs to be kept in mind that the push in Asia to develop bond markets and undertake other financial reforms is relatively new and there is a limit as to how quickly things can change. The relevant comparison therefore is not with markets in mature economies today, but with the situation when those markets were at a similar stage of development as Asian markets are today. Take, for example, the question of market turnover. While the turnover ratio in many Asian bond markets is well below that in Australia today, it is very similar to that in Australia in the years that followed the introduction of bond tenders in 1982. I therefore do not think that there is any reason to believe that Asian bond markets are intrinsically less liquid than those elsewhere; it is simply a question of their having sufficient time to develop.
Aside from criticisms about the markets themselves, some commentators have also questioned the effectiveness of some of the initiatives that are being pursued, particularly the Asian Bond Fund. I have seen three types of criticisms:
The first is a general concern about ‘moral hazard’ from public sector involvement in setting up and investing in local bond funds. The particular concern seems to be that there may be a temptation to use these funds to bail out individual issuers or to manipulate market prices.
I do not attach a lot of significance to this criticism. The funds have been set up in a way that limits the power of any individual country authority to interfere in the running of the fund with a view to advantaging its own issuers. The criteria for selecting bonds to be included in the benchmarks against which the funds are managed need to have the support of all the countries involved, and the people at the individual central banks who are accountable to their own authorities for the funds invested would strongly resist any actions which would increase their risk of financial loss.
A further safeguard is that the funds themselves are managed by independent managers – the BIS in the case of ABF1 and private sector managers in the case of ABF2.
A second criticism is that the setting up of these funds will do little to improve
local bond markets, and could in fact damage them. There are various arguments
- the amount of money invested is very small relative to the size of the markets and therefore will not have much impact;
- central banks are largely ‘buy and hold’ investors who will not add much liquidity to the market; and
- that by focusing mainly on the highly rated segment of bond markets, where there is already plenty of demand, these funds could crowd out private sector investors.
I think we can all readily accept that investing $1 billion or even several billion dollars in a market that is capitalised at around $1,400 billion is not, in itself, going to have a lot of impact on market development. By the same token, of course, it is not going to have any crowding out effects.
I would also readily accept that these funds are not going to add significantly to market turnover. They are simply too small to do so, and are being managed relatively passively against a defined benchmark.
Nonetheless, I think these criticisms miss the point. In setting up these funds, the aim of the participating central banks was not to promote bond market development by adding directly to demand or turnover; rather, it was to help identify, through their direct involvement in markets, the barriers that exist to market development. The benefit to markets was intended to come not from the investment itself but from policy reforms that individual national authorities might take. Doing this on a collective basis has had the added advantage of information sharing and creating peer pressure to maintain the momentum of the reform process.
- The third type of criticism leveled at the ABF initiatives is that the efforts to integrate the various national bond markets in Asia, or harmonise regulations and taxation, will encourage more cross-border capital flows and are thus the equivalent of capital account liberalisation. It is argued that liberalising the capital account in the absence of strong and well-developed markets may entail ‘sequencing risks’, in that regional financial markets may not yet be in a position to handle increased volumes of cross-border flows.
It may well be the case that the integration of bond markets amounts to capital account liberalisation, but I don't think this should stand in the way of market development. Whatever the theory might say about sequencing of reforms, in practice the reform process must take opportunities as they arise and often has to proceed on many fronts at once as reforms in one area rapidly result in pressures for change in other areas.
In any event, the argument that bond market reform could cause problems for exchange rate management is, I feel, overstated. It attaches too much importance to cross-border flows in the development of local currency bond markets. Such flows in relation to bonds are small relative to both the size of the bond market and to cross-border flows arising from other types of investment. Even if the various initiatives did result in a large increase in cross border flows, it is unlikely that such flows would become a significant force in foreign exchange markets in the foreseeable future.
In summary, I am optimistic about bond market developments in Asia. Markets have already come a long way since the Asian crisis. I accept that, on many measures, Asian markets still lag behind those of the mature economies but such comparisons should be seen as highlighting the potential that exists, not as a criticism of the current state of markets. It would be unrealistic to expect that Asian markets could have made up, in just a few years, the ground that markets in mature economies took decades to cover. The co-operative approach being taken in the region, by allowing countries to benchmark off each other, will ensure that a rapid pace of reform will continue. In turn, if I may end by relating this back to the theme of this morning's session, I think this will create a great deal of opportunities for financial service providers in the region.