- You cannot purchase the bond directly on-line on this particular Web site. As is the case with many discount brokers, the Vanguard brokerage firm, on whose Web site the listing appears, does not own the bonds being offered. Instead, Vanguard is using a database (Bond Desk) that lists offerings made by a large number of dealers who actually own the bonds. To actually buy the bonds, you need to call Vanguards brokerage desk.
- The difference between the bid and the offer price, when listed, represents the commission charged by the listing broker. The broker in this example, Vanguard, also adds a small fee to the offer price ($3.00 to $5.00 per bond depending on the size of the trade), so that the actual yield will be slightly lower than the offer yield.
- The yields of high-quality credits such as Australia and Quebec are roughly in line with those of domestic corporate bonds with comparable credit quality, maturity, and call features. The higher the yield, or the spread to U.S. Treasuries, the higher the risk. Aside from the bonds of Argentina, which are in default, the highest yield listed is that of Mexico, at slightly above 8% for bonds maturing in 29 years. In an environment of rising domestic rates, you have to seriously consider whether you are being compensated for the high risk involved in both the weak credit quality and the long maturity.
- Global bond funds. These invest in the bonds of at least three countries, one of which may be the United States.
- International bond funds. These invest mainly in bonds of foreign issuers. Those tend to be foreign-pay bonds.
- Emerging markets bond funds. These bonds funds invest mainly in the debt of emerging markets. Those are also called developing or underdeveloped countries.
- Short World, Multi-Market Funds. Again, these also invest mainly in foreign-pay bonds. They have two distinguishing characteristics: Maturities are typically under two years, and they use derivatives to hedge currency exposure.
How to Buy International Bonds and Funds
by Annette Thau
This is the second article of a two-part series on investing in international bonds. The prior article discussed the types of bonds available, the distinctions between international and emerging markets bonds, and some of the history of both markets (Foreign Interest: A Closer Look at the International Bond Markets, May 2004 AAII Journal).
This article will discuss a number of different options for actually investing in international bonds, including the purchase of individual bonds as well as mutual fundsthe most practical approach for individual investors. Well also take a look at past returns.
As the prior article pointed out, one of the main distinctions between international bonds is whether they are denominated in U.S. dollars or in a currency other than the U.S. dollar (these are called foreign-pay bonds).
Buying Bonds Denominated in U.S. Dollars
International bonds denominated in U.S. dollars can be purchased from brokerage firms selling taxable bonds. Such bonds are also listed on Web sites offering bonds for sale and are found under the heading of Corporate Bonds. (On the Web site I use most oftenVanguardyou click successively on Corporate Bonds, Secondary Market, and then Foreign Issuers.) There is an extremely broad variety of offerings, in all maturities, with credit ratings from the very highest quality (AAA) to bonds well below investment rating or in actual default (D rating).
Table 1 offers a small sample of what the listings will show; this sample listing appeared on January 22, 2004. Additional details can be found about each bond by clicking on the name of each issue on the Web site.
The listing is mainly self-explanatory, but a number of terms may be unfamiliar. These are explained in the sample listing in Table 1. You should also note the following:
|TABLE 1. Secondary Market: A Sampling of Foreign Issuers|
|1775||Quebec Province Can||7.5||9/15/2029||120.32||5.94||A1/A+|
|250||Euro Bk Recon Dev||FLT||7/22/2015||90.15||NR/AAA|
|250||United Mex Sts Mtn Be||7.5||4/8/2033||96.75||8.58||Baa2/BBB|
|100||Argentina Republic||12.25||6/6/2018||29.50 Trading Flat||41.94||CA/D|
The Listings: What the Terms Mean
FLT stands for Floater, which indicates that the coupon is reset periodically according to a specified benchmark interest rate. This is a very common feature of foreign bonds.
The term Trading Flat, which accompanies the Argentina bond listing, indicates that the bonds are not paying interest. The indicated yield of 41.94% is not a true yield since the bonds are not paying interest. The D credit rating indicates that the bonds are in default. Obviously, this is a highly speculative offering. Any bond selling for approximately 29 cents on the dollar is equally speculative.
A typical listing will also include the bonds identifying CUSIP number (left out here for space considerations). It may also contain a bid price if available to the listing agent, and a bid yield. For example, the bid price for the Italy bonds in the above listing was 91.83; the bid yield was 5.97. The difference between the bid and the offer price, when listed, represents the commission charged by the listing broker.
Source: Vanguard Brokerage.
Why might you want to buy international bonds denominated in U.S. dollars?
Two possibilities. One would be to speculate by buying bonds with very high yields in markets that have potential for appreciation. (Brady bonds were all initially dollar denominated and rated as the lowest form of junk; ultimately they proved to be outstanding speculations.) But of course, many highly speculative bonds are losers.
At the other end of the spectrum, since the yields of higher-quality credits are roughly in line with those of U.S. corporate bonds, such bonds would be of interest mainly to individuals who want to diversify an extremely large portfolio of taxable bonds.
Buying Foreign-Pay Bonds
Buying foreign-pay bonds is far more difficult than buying U.S. dollar-denominated bonds. I called a number of large international banks, as well as brokerage firms that have international operations.
The response I got from Merrill Lynch was typical: If you are considered a high net worth client of the firm, your broker can obtain foreign-pay bonds for you. But you have to show a specific need and be very clear about what you want and why. It is the policy of Merrill Lynch to discourage (or, as the broker put it, not to encourage) individuals from buying foreign-pay bonds. Rather, the official policy of Merrill Lynch is to suggest that if you would like to invest in foreign-pay bonds, you should do so through bond funds. Exceptions would include foreign nationals living in the U.S. who might want to invest in currency of their own country, or U.S. citizens doing business abroad. In such cases, Merrill would put you in touch with a branch office in the country of interest. Similarly, if you are a wealthy citizen of a Latin American country, a bank such as JPMorgan Chase can service your needs. But, in most instances, at the current time, individual investors will find buying foreign pay bonds extremely cumbersome.
Other than bond funds, one other alternative is foreign-denominated CDs (for more on this, see the accompanying article).
Because investing in any type of international bond requires genuine expertise, individual investors will find that bond mutual funds represent the most practical way to invest in such bonds.
International bonds are extremely diverse. Offerings range from corporate and sovereign bonds with very high quality credits from highly developed countries, to struggling firms in countries with struggling capital markets.
This, in turn, results in an enormous variety of options for bond funds putting together a portfolio of international bonds.
As a result, international bond funds vary a good deal in their makeup. Since the total returns reported in the major financial papers are generally based on data published by Lipper Inc., the Lipper categories are a good place to start. Lipper broadly distinguishes between four categories of international bond funds. These are:
|TABLE 2. International Bond Funds: Total Returns (Through 5/31/04)|
|Cumulative Return (%)||Annualized Return (%)|
|3 Mos||YTD||1 Yr||3 Yrs||5 Yrs||10 Yrs|
|Short World, Multi-Market||-1.63||-0.76||1.19||6.02||4.20||4.18|
|Source: Data supplied to the author by Lipper Inc. Data as of 5/31/04.|
Table 2 shows total returns for these four categories of funds through May 31, 2004. Returns are cumulative up to one year, and annualized for periods over one year.
As Table 2 shows, annualized total returns in all four categories are much higher for the most recent three- and five-year periods than 10-year returns. Since the beginning of 2004, however, through the end of May, returns have headed south. The interesting question is: What lies behind these returns? For some answers, lets briefly look at each category.
International and Global Bond Funds
These two categories include, by far, the largest number of international bond funds.
To analyze the factors behind the returns, lets look at the year-to-year returns of one international bond fund: the T. Rowe Price International Bond Fund, one of the oldest international bond funds. It is also a no-load fund and is conservatively managed. Table 3 shows returns for the period beginning in January 1993 and ending December 31, 2003.
Table 3 also shows the components of the returns. The income return consists of the dividend payments made by the fund, as well as interest earned by reinvesting those dividends. Together, those add up to the funds yield. The capital return represents the capital gain or the capital loss experienced by the fund as a result of changes in net asset value (NAV). Adding these two cash flows together gives you the funds total return for the year.
|TABLE 3. T. Rowe Price International Bond Fund: Year-to-Year Returns, 1993 Through 2003|
|Income Return (%)||7.36||5.95||6.86||5.92||5.16||5.47||3.76||4.39||3.90||3.25||2.78|
|Capital Return (%)||12.67||-7.79||13.45||1.22||-8.33||9.56||-11.62||-7.52||-7.31||18.55||16.00|
|Total Return (%)||20.03||-1.84||20.31||7.14||-3.17||15.03||-7.86||-3.13||-3.41||21.80||18.78|
For domestic bond funds, NAV goes up or down chiefly in response to interest rate changes. That is true of international bond funds as well. But changes in the value of the dollar against foreign currencies also play a role in determining total return: When the dollar is weak, the NAV of foreign bond funds goes up; when the dollar is strong, it declines. In fact, foreign currency moves relative to the U.S. dollar play a larger role than interest rate changes in determining total return.
This is very clearly illustrated in Table 3. Take a look at the returns shown for 1993 and 1994. Average yield for all of 1993 was 7.36%. That yield declined to 5.95% in 1994, a decline of 141 basis points. In a domestic bond fund, a decline of this magnitude would result in a sizable capital gain. But instead, 1994 shows a capital loss of close to 8%. The reason for the loss is that a steeply rising dollar swamped the decline in yield, resulting in a capital loss and a negative total return for the year. The story is the same for several other years, including 1997 and 1999. For example, 1999 is particularly striking because between 1998 and 1999, the yield of the fund declined from 5.47% to 3.76%a decline of 171 basis points. But again, the rising dollar resulted in a capital loss of almost 12% and a negative total return for the year.
Just the opposite occurred in 2002 and 2003. For both of those years, yields were very low, hovering around 3%. But a declining dollar resulted in strong capital gains for each of those years (18.5% and 16%, respectively), and very high total returns (22% and 19%, respectively). Since the beginning of the year, however, a strong dollar has so far reversed this positive trend. If you go back to total returns shown in Table 2, the data shows average total returns that are somewhat higher for the international group than for the global group. This might be due to the fact that, according to Lippers definition, the international group excludes investments in U.S. bonds. As a result, exposure to currency fluctuations against the dollar may be higher for this group than for the global category. In an environment where the value of the dollar is declining, funds with the greatest exposure to currency changes would have the best performance. But the lines between the two groups are fluid. Both the global and the international categories have strong and weak funds.
If you use Morningstar to research funds, you should note that Morningstar does not differentiate between international and global categories. All are lumped in one group.
Emerging Markets Bond Funds
The emerging markets bond funds are a much smaller group than either the international or the global groups. The chief difference between funds in this category and other international bond funds is one of credit quality. These are the junk bonds of the international bond market. Credit ratings of bonds in this group are determined chiefly by geopolitical risk, another term for unstable governments and unstable economies. This group, however, has been the star of the international bond market. What is more, total returns of the emerging markets group have been high for the last five years, longer than any other international bond fund group.
To analyze the factors behind these returns, lets take a look at year-to-year returns for one emerging market bond fund, the T. Rowe Price Emerging Markets Bond Fund. This fund had its inception relatively recently, in 1995, but it is still one of the oldest funds in this sector. It is also conservatively managed and has no load. Table 4 shows returns for the period of 1995 through 2003, and includes interest income and capital gains.
|TABLE 4. T. Rowe Price Emerging Markets Bond Fund: Year-to-Year, Returns 1995 Through 2003|
|Income Return (%)||na||na||10.81||9.94||9.22||10.22||12.39||10.83||11.55||8.51||7.46|
|Capital Return (%)||na||na||15.05||26.85||7.63||-33.30||10.58||4.37||-2.21||1.01||18.59|
|Total Return (%)||na||na||25.86||36.79||16.85||-23.08||22.97||15.2||9.34||9.52||26.05|
Compare Table 4 to Table 3. Note, first of all, the income yield for the years between 1995 and 2001. During all those years, it was close to, or above, 10% (above 12% in 1999). These very high yields were a significant factor in the total returns of those years. High yields were tied directly to the perception that credit quality was poor and the risk of default high. Indeed, disaster struck in 1998 (the year of the Asian currency crisis), which resulted in a capital loss of over 33%, and a net loss for the fund of 23%in spite of the high yield.
But between 2001 and 2003, a veritable sea change took place. There was, first of all, a dramatic improvement in credit quality, as the credit rating of bonds of a number of developing countries went from junk to investment grade. At the same time, as interest rates in the U.S. declined to historic lows, bonds with higher yields and improving credit quality proved to be irresistible to many buyers who previously shunned these markets. Both of these factors eventually resulted in significant capital gains (almost 19% in 2003), but also much lower yields.
|FOREIGN CURRENCY CDs|
If you would like to hedge against a declining U.S. dollar, or speculate on foreign currencies, an on-line bank named EverBank (www.everbank.com) offers an interesting option: CDs denominated in foreign currencies. These are called World Currency CDs. This product actually has a fairly long history. World Currency CDs were first offered in the 1980s by a small bank named the Mark Twain Bank, which was purchased by EverBank in the year 2000. EverBank is strictly an on-line bank.
Like many CDs, World Currency CDs are offered for periods of three months, six months or one year. The CDs are insured by the FDIC. What is unique about these CDs, however, is that they are denominated in foreign currencies. EverBank offers a broad variety of options. The first is to invest in a single currency. Currently being offered are CDs denominated in major currencies, such as the euro and the Japanese yen; the currencies of smaller countries, such as the Czech koruna and the Danish krone; and more speculative currencies, including the Mexican peso and the South African rand.
Another option is to buy a basket of currencies tied to a particular theme. At the present time, EverBank offers several choices. One, called the commodity CD is tied to currencies of countries that produce commodities: Australia, New Zealand, Canada and South Africa. Another, tied to gasoline, is called the petrol CD. Soon to be introduced is a prudent CD, which will invest in a basket of strong currencies.
The required minimum investment is $10,000 for a CD denominated in a single currency; and $20,000 for one denominated in several currencies.
Interest is paid on these CDs in the foreign currency. A few CDs offer significantly higher interest rates than are currently available on more conventional CDs. The highest are denominated in South African rands (7.45% for a three-month CD) and the Mexican peso (4.84% for a three-month CD). Obviously, these are speculative currencies.
CDs denominated in euros or Japanese yen currently pay extremely low interest rates: 0% for a one-year Japanese CD, 0.5% for a one-year euro CD.
Why, you might reasonably ask, would anyone want to invest in any of these CDs?
Well, the answer for both high and low interest rate CDs is that total return is governed primarily not by the interest earned on the CD, but rather by the movement of the currency against the U.S. dollar. The FDIC insurance does not protect against losses to the market value of the CD. If the dollar rises against the foreign currency, the CD suffers a capital loss. Of course, if the dollar declines against the currency, the CD earns a capital gain. Currency movements can be steep and are highly unpredictable. Therefore, volatility of returns can be high.
EverBank does not publish total returns. But a chat with a friendly bank officer gave some indication of what happens if you choose a CD denominated in a currency that is volatile against the U.S. dollar. The top performer in 2003 was the Australian currency. If you had purchased a one-year CD for 2003, the CD appreciated by 34% (capital gains) in addition to interest earned. But in 2001, the Australian CD experienced a capital loss of 14%. For the five-year period between 1996 and 2001, a period when the dollar was very strong, the Australian CD declined by 37%. Even after adding back interest for that five-year period, the CD experienced a net decline of 17%.
In addition to the plain-vanilla CDs, EverBank also offers CDs denominated in foreign currencies but intended for enterprises that do business in foreign countries. These are designed for businesses that want to hedge or lay off some of the currency risks inherent in doing business abroad. The cost of using the CD is lower than hedging using options or futures.
Larger banks with international branches such as Credit Suisse, First Boston and JPMorgan Chase, offer similar CDs, but require much higher minimum investments of $100,000 or higher.
At the present time, the yield for this fund is around 6%. (Note that while this particular fund had only very modest capital gains in 2002, 2002 was a standout year for other emerging markets bond funds.)
In my prior article, I cautioned that future returns of emerging markets bond funds were unlikely to match those of the past five years. Why this conclusion? Several reasons. One is that yields in many of the emerging markets have declined to a point where the spread to U.S. Treasuries has reached historic lows, and the narrow spread makes these bonds much less attractive. The second reason is that a widely anticipated rise in U.S. rates is likely to cause losses in many sectors of the bond market, with higher losses coming in weaker sectors. So far, 2004 has followed this script. At the beginning of the year, emerging bond markets started up, but they have since turned down and are either very modest or negative for most emerging markets bond funds.
Short World, Multi-Market Funds
The short world, multi-market category has the fewest number of funds, and the smallest amount of assets. It consists of a small group of funds, introduced with great fanfare in the early 1990s. At the time, interest rates were higher in foreign markets than in the United States. The funds were marketed as low risk but higher yielding alternatives to money market funds. Initial returns were high because initial interest rates were high, and also, because the dollar was sinking. Subsequently, as interest rates declined and as the dollar rose, returns have ranged from very modest to very poor.
If your objective is to keep your money safe, in a low-risk investment, this category is a poor choice. Note also that, as is true of other international bond funds, returns vary significantly from fund to fund. Average total returns for the entire group are boosted by the returns of one fund with an annualized total return of 7%. Annualized total returns for longer periods for most funds in this group are around 1% or 2%lower than those of riskless U.S. money market funds.
Lets assume you want to invest in an international bond fund. What criteria would you use? There is no easy answer. Returns vary significantly from fund to fund and from year to year. As an example, for 2003, in the two largest categories, international and global, the yield of individual funds varied between 2.5% and 9%, with some outliers above 9%. Annualized total returns for the preceding five years covered an even wider range, from 5% to above 20%. Keep in mind, however, that average total returns are somewhat skewed by the fact that few funds have been in existence for as long as 10 years.
Note also that a funds strategy with regard to hedging is a key factor in determining total return. But the same strategy has variable results depending on the direction of currency changes. In 2003, when the dollar declined significantly against many major currencies, funds that did not hedge currency risk had much higher total returns than funds that did hedge. To date, in 2004, the dollar has been strong. This gives the advantage to funds that hedge.
Similarly, in the emerging markets group, portfolio managers have to be astute (or lucky) enough to pick the markets that are going to do well that year. No manager consistently picks winners year after year.
Before deciding to invest in any international bond fund, it is a good idea to look at the funds actual returns from year to year. Many funds post what look like solid returns for, say, a 10-year or a five-year period on an annualized basis. But as shown by the two T. Rowe Price funds, long-term annualized returns can mask a lot of volatility. When you look at performance from year to year, you may find that the annualized returns result from two years of strong returns, two years of disasters, and modest returns in other years.
Finally, fund managers often change strategies in ways that significantly affect fund returns, for good or bad. In 2003, for example, two of the best performing international bond funds boosted returns by investing a large portion of fund assets in the bonds of emerging markets. Some managers hedge currencies one year and not the next.
Note also that fund expenses tend to be very high in this grouptypically, between 1½% and 2%. Also, a number of the better-performing funds, at least according to Morningstar, are loaded funds. If a fund has a 5% load, at current interest rates you are giving up a years worth of income the first year you invest. In addition, many of these funds have expense ratios from 1.25% up to 2% a year. Lets do a back-of-the-envelope calculation for a two-year return: Assume that the portfolio earns 10% per year for two years, or a total of 20%. Subtract a 5% load and 3% for expenses. After two years, net return to you: 12%, or 6% a year. But even that overstates returns because it does not include taxes on dividends or capital gains.
If you like to speculate, then this is an interesting alternative to open-end mutual funds.
Lipper lists only two categories of international closed-end bond funds: international and emerging markets. The makeup of the portfolios of these funds is similar to that of their open-end cousins. A number of well-known mutual fund groups manage both closed-end and open-end international funds (for example, Pimco, Templeton, and Alliance). But there are far fewer closed-end funds than mutual funds, with correspondingly smaller amounts of money under management.
Closed-end funds differ from their open-end counterparts in that the shares are not continuously issued by the fund but instead sell like stocks on the various stock exchanges. The price of shares changes throughout the trading day and commission costs are in line with those of any stock.
|TABLE 5. Closed-End International Bond Funds: Total Returns (Through 5/31/04)|
|Cumulative Return (%)||Annualized Return (%)|
|3 Mos||YTD||1 Yr||3 Yrs||5 Yrs||10 Yrs|
|Data as of 5/31/04.|
Closed-end funds are trickier to analyze than open-end funds because of their structure. Unlike their open-end cousins, closed-end funds do not sell at the net asset value of the assets in the fund. Typically, they sell at a discount to net asset value, but that can move to a premium if the funds are in high demand. In addition, many bond funds are leveraged. What that means is that they borrow money to buy additional bonds (that is, they buy on margin) in order to boost yield. The leverage often does result in a higher yield. But leverage also magnifies volatility, both up and down. Average returns for the entire group are slightly higher for closed-end funds than for open-end funds (Table 5). But again, the averages mask significant differences from fund to fund. Annualized total returns vary from 6% to 12% for individual funds.
If you are considering investing in closed-end funds, be prepared for significant volatility. If interest rates rise, then the funds suffer a triple whammy. First, the price of individual issues in the portfolio declines as a result of the rise in interest rates. But in addition, the discount widens. And for leveraged funds, the interest cost of borrowing on margin rises. In 2004, as interest rates rose, many closed-end bond funds suffered declines of 10% in a matter of weeks.
While returns of international bonds have been very good over the last three to five years, future returns are not predictable.
Returns for the best-performing sector, emerging markets bonds, are unlikely to match those of the last five years because yields in this sector are at historic lows.
Returns for higher credit quality international bonds will depend primarily on the direction of the dollar against major currencies. A weakening dollar would result in higher returns. A strong dollar would mean lower returns. But a rising interest rate environment in the U.S. would cause erosion in all markets, and particularly in emerging markets.
The rationale for investing in international bonds is grounded in modern portfolio theory. It holds that even though this is a risky asset class, it lowers the total risk of the portfolio because returns do not correlate with those of U.S. bonds.
But, in my view, this argument is valid mainly for individuals with extremely large bond portfolios who want to diversify bond holdings. This argument, moreover, is theoreticalthere is not yet sufficient historical data to back up the theory.
If you have a very large bond portfolio, you might choose to invest a small portion in the international area. However, for individual investors with more modest holdings, this is not a must-have category.
Ms. Thau is a former municipal bond analyst for Chase Manhattan Bank. She also until recently was a visiting scholar at the Columbia University Graduate School of Business.
Annette Thau, Ph.D., is author of The Bond Book: Everything Investors Need to Know About Treasuries, Municipals, GNMAs, Corporates, Zeros, Bond Funds, Money Market Funds, and More, (copyright 2001, published by McGraw-Hill; $29.95). She has spoken to AAII chapters in different parts of the country about bonds and bond funds.
Ms. Thau is a former municipal bond analyst for Chase Manhattan Bank. She also until recently was a visiting scholar at the Columbia University Graduate School of Business.