Prescription for the Indonesian bond market

By Thads Bentulan

HONG KONG (JP): There is a perfect reason why the writer is playing doctor to the Indonesian bond market -- there exists no rupiah government benchmark.

Now, this could not be a perfectly correct statement because there actually exists a debt instrument called Sertifikat Bank Indonesia (SBI) and still another called the bank recapitalization (Recap) bond but I am going to demolish their standing as possible useful benchmarks later.

If you scan the financial pages of any regional newspaper, any listing of an Indonesia benchmark is invariably the missing.

Why is there a need for a benchmark? Benchmarks are so important, that I am of the conclusion the Hong Kong Monetary Authority sells treasury bills called exchange fund notes primarily to establish benchmarks rather than to raise funds. After all, Hong Kong perennially has surplus budgets; therefore, it does not need to borrow.

A government benchmark is needed to give bond investors a realistic and computable assessment of risk. Without a "risk- free" reference parameter given by the sovereign, the relative risk profile of ordinary entities like corporates or banks cannot be quantified. If risk cannot be quantified, learned investors may avoid the market altogether.

Around Asia, the government bond issuances are very high in South Korea -- an OECD member -- comprising about 47 percent relative to the gross domestic product (GDP). Indonesia's government bonds attained 45 percent relative to GDP, while Malaysian government bonds reach as high as 25 percent. These are all 1999 figures.

In the corporate bond markets, Korean corporates raised about 25 percent relative to GDP, Indonesia corporates raised just 1 percent while Malaysian corporates raised 26 percent.

Last year, Korea issued about US$290 billion in domestic currency debt, followed by China with $155 billion and Malaysia with almost $70 billion. Indonesia issued around $12 billion in local currency bonds.

Zeroing in on Indonesia, the government bonds are comprised of 36 percent in floating rate, mostly recap bonds, also about 36 percent in inflation-indexed bonds, and 21 percent in fixed rate bonds.

Corporate issuers raised more than Rp 5 trillion ($523 million) in the first semester of 2000, which is not bad compared to last year's Rp 2.2 trillion ($230 million).

Of the outstanding corporate issues in the market, about 37 percent are from the manufacturing sector, followed by the financial sector which comprise 26 percent, then by the real estate sector, and also the utilities sector with 16 percent share each.

While profiling the bond investors in Indonesia is difficult, this much we know: Pension funds hold only about 6 percent of their portfolio in bonds way below the regulatory limit of 20 percent.

Insurance companies only carry about 4 percent in bonds.

Mutual funds are hard to pin down because a recently introduction tax regulation necessitates a liquidation of the investment very five years.

As for the banks who are the traditional bond investors elsewhere in the world, they park their excess funds in the SBI, which of course is a short-term instrument.

The first problem is the lack of a yield curve. Right off, the absence of a yield curve tells you there is a problem with Indonesian bond market.

The SBI is a short-term rupiah debt instrument issued by the central bank, the Bank of Indonesia. With maturities ranging from only one to three months these SBI provides a benchmark for the shorter end of the yield curve. That is, the SBI is a useful benchmark only for working capital loans but useless for bond issues.

In addition to the benchmarking function, the primordial function of government bonds, of course, is to provide long term investment instruments for the investors with natural long term liabilities. I am of course referring to insurance and pension fund companies.

From the point of view of the government, the SBI is fundamentally a monetary intervention instrument used to absorb excess liquidity from the banking system when needed. But since the government does not use the SBI to finance budget deficits and government expenditure, by analogy, corporate borrowers cannot use the SBI to benchmark their own capital expenditure borrowings.

By the way, in the recent months, at times the SBI's yield are higher than the longer term recap bonds, a seldom seen yield curve inversion.

The Indonesian Bank Restructuring Agency (IBRA) which was created following the 1997 currency crisis, acquired the non performing loans of the ailing banks and issued about Rp 500 trillion ($52.3 billion) in bonds to pay these assets. These are the so-called recapitalization bonds.

Of these recap bonds, some were 12 percent fixed rate five- year bonds, some were 14 percent fixed rate 10-year bonds, and there were 16 series of floating rate bonds with various maturities ranging from due 2002 bonds to due 2009 bonds.

Why are these bonds unsuitable as benchmarks?

First, they were issued at par, instead of being priced against the local bond markets.

Second, since there was no treasury auction such as those done in the U.S., the value of these bonds are but nominal value, therefore, they do not reflect the sovereign risk of Indonesia from the viewpoint of the bonds investors.

Third, the banks under IBRA are holding these recap bonds in their balance sheets as investments not as trading instruments, therefore there is no secondary market liquidity.

Although these banks are now allowed to trade a maximum of 25 percent of their recap bonds, they hesitate to do so. Banks are hesitant to trade these bonds because the interest rates are picking up, therefore they have to be sold below par.

When that happens, the banks will suffer a capital loss because the trading portfolio will be marked to market. In short, by moving their recap bonds to the trading portfolio, banks are risking another balance sheet loss problem.

Fourth, from the very start, the recap bonds were more of a balance sheet cosmetic: That is, recap bonds were exchanged for document-based non-performing loan assets rather than being exchanged for cash from investors.

Fifth, together with the above points, the recap bonds were never intended to be used a benchmark unlike in other countries where governments aim for the twin goals of fund raising and benchmarking.

When government regulations classify tax regimes based on the type of investor, rather than the type of instrument, the development of the debt market eventually suffers. This is the case with Indonesia, especially regarding the classification of on-shore and off-shore investors.

Indonesian offices of foreign banks are still not allowed to issue long term rupiah bonds which again stifles debt market development. This is unfortunate because foreign banks are often highly capitalized to back up their rupiah liabilities.

Approval process is slow, which is not uncommon in Indonesia, therefore the government should introduce the shelf approach to bond issuances similar to Rule 415 in the U.S. This solution eliminates the repeated approval process each time there is a take down. In short, we solve the problem by bypassing it.

The medium term note programs established by a few Indonesian conglomerates for overseas borrowing is a good model to follow.

Since there are no treasury auctions, there are no primary markets or dealers to speak of. By setting up the systems of regular treasury issuances, primary dealers can be appointed.

The lack of a yield curve, of course, creates pricing problems for corporate bond issuers who eventually resort to curve interpolation, which again, reduces the reliability of the risk profile of the private entities as against the sovereign.

The lack of a government benchmark gives rise to too many confusing proxy benchmarks which ultimately redounds to the mispricing of corporate issuer risk.

The lack of a common clearing system for the both government and private sector bonds can be solved by hiring consultants from Euroclear or Cedel to implement one for Indonesia.

Leaning towards book-entry bonds, rather than physical delivery of bonds, should be a primary goal of the local debt market system. The government can take the lead role in this endeavor.

Credit ratings are important and will surely follow the development of the market.

Of course, taxation is important. If the intention is to develop the local debt market, then tax incentives for debt instruments should be used to pull investors away from other securities. For example, capital gains taxes or final taxes can be changed to favor bonds.

Altogether, the first step is for the government to provide the necessary debt instruments that covers the entire range of tenors. Once the tenors are established, the yield curve will reflect the sovereign risk that will then encourage investors to quantify corporate issuer risk.

The writer is a Hong Kong-based head of research of a market research house and columnist for Manila-based BusinessWorld newspaper. The above is based on his article in a Hong Kong capital markets magazine (His email: streetstrategist@hotmail.com.)