nick dewhirst casts a suspicious eye over EM bonds asking how long their recent success can last and what the warning signs will be
In recent times, emerging market bonds have been the hottest game in town. While stocks have slumped over the past three years, many emerging market bond funds have appreciated by as much as 30%.
In the first two months of this year, this sector attracted around $400m new subscriptions in the US. That is triple the amount at this time last year, according to a recent survey of 200 funds managing some $9bn.
I am a great fan of emerging market bonds. I have personally bought one 5% Sterling government bond issued by China and another by Russia, as well as a 3% undated Frank bond issued by Turkey. They make fine decorations and are an excellent reminder of just how little faith should be placed on the binding engagement of such governments and their successors, for they are all worthless.
Whenever an investment game has become exceedingly popular, it is wise to ask if it is not also becoming excessively popular. The simple truth is that the better performance in the recent past, the greater the popularity, but less performance can be expected in the future, because so much of the potential has already been exploited.
Here, therefore, are the potential pitfalls for investors considering playing this particular investment game at such a late stage.
There are few developing countries that are strong enough to have active domestic long- dated bond markets. Korea, Poland and South Africa are among those few. Most corporate or government bonds are denominated in the world's principal reserve currencies, mainly the US dollar but increasingly also the euro.
There are historical reasons for this. The market came into existence with the launch of Brady Bonds, named after the US Treasury Secretary responsible for creating marketable securities to consolidate the debts of recovering Latin American governments a decade ago.
Today, the market is worth some $200bn and includes over a dozen governments in Asia, Africa, Eastern Europe as well as Latin America, but they are still all denominated in US$.
As a result, these funds are essentially US dollar-denominated investments. The Emerging Markets Bond Fund managed by M&G is a typical example. At the latest report, its assets were invested are 94% in US dollars and 6% in euros.
Thus, any rally in the exchange rates of many emerging markets does little or nothing for these funds. For example, the Brazilian Real has appreciated by 20% against the US dollar so far this year, but international investors have not benefited from the currency movement because most of Brazil's Brady bonds are denominated in US dollars.
The position is worse for investors whose domestic currency is the euro, because they will experience capital losses. The euro has already made good its 15% decline since launch, and is now hitting new all time highs.
The biggest profits are made in bond markets when inflation declines, for that pushes yields down and bond prices up, but much of that has already happened.
US inflation peaked at 15% in March 1980. A year and a half later the benchmark US long bond also peaked at that level. Today, US inflation is running at 3% and the bond yield has fallen below 4%.
Emerging market bonds have reflected this trend. Yields around 20% were obtainable at several stages in the past decade, but now yields have halved, as can be seen in chart one.
As a result, the prices of many US dollar-denominated sovereign bonds have risen to levels where capital losses are guaranteed. Turkey's 11.75% 2010 bond trades at 106%, South Korea's 8.875% 2008 trades a quarter above its redemption value, and Mexico's 11.5% 2026 bond trades more than 50% above that.
Even where bonds are denominated in local currencies, the biggest gains are history. Taking Korean Won bonds as an example of one of the largest and finest emerging market bonds, chart two shows how the decline in Korean inflation has been reflected in falling yields.
As a result, not only are the South Korean government's US dollar-denominated bonds priced to guarantee capital losses, but so also are many of its domestic currency bonds.
Not only have yields shrunk due to falling inflation, but also due to improving confidence. Brazil has not followed its southern neighbour into populist economic policies and Russia has even repaid a Eurobond. The sector, as a whole, is basking in the glow of such encouragement by these two major participants.
Again, taking Korea as our example, yields have converged. In 1991 yields on Won bonds peaked at 17%, which was 10% more than US Treasuries at the time. Now the yield has shrunk to little more than 4% and that is less than half of one per cent more than US Treasuries.
This trend is reflected across the entire developing world. The EMBI+ Composite Yield spread has also shrunk dramatically. Since 1990 it has ranged from 3% to 11%. Now it is near the middle of the range at 6%.
This convergence is also reflected in the decline of real yields after inflation. Back in 1991, the nominal Won yield was nearly 8% higher than the rate of inflation. Now that real yield has also shrunk to less than half of one per cent.
Given the strength of the Korean economy and the weakness of the US, it is conceivable that treasuries and Won bonds could even trade places, but the scope for further declines in yield must be severely limited, given the low level of both at this stage.
The curse of closet-index linking affects this sector like every other. Fund managers who stray from benchmark weightings lose their jobs if they get it wrong, but can blame it on unanticipated market conditions if they copy their peer group.
Indeed the curse is probably worse for junk bond funds, of which this is a sub-sector. Like equity indices, bond indices are weighted by market capitalisation. Unlike equity indices, bond indices are therefore weighted by the size of each country's debts. Thus the biggest debtors get the biggest weightings, so forcing the closet-indexers to invest in the most indebted, irrespective of the wisdom of that strategy.
There have been two recent examples of this problem. Russia was over-represented before the rouble default in 1998 and Argentina accounted for as much as a quarter of some indices, before its debt crisis in 2001.
Investors checking their fund's country weightings after such disasters may feel relief at seeing how little of their fund's assets were still accounted for by investments in such countries. However, they delude themselves. It is unlikely that exposure has fallen as a result of prudent sales, because such bonds then become illiquid. It is more likely that exposure has fallen because the bonds have been written off.
While there may well be special situations in the local currency bonds issued by some countries like South Korea or South Africa, it is unlikely that these feature highly in any of the benchmarked funds.
These concerns may be premature. After the succession of currency crises in the past five years, many emerging markets have cleaned up both their corporate and public sector balance sheets. If the world economy now recovers, this situation will further improve. However, it may re-ignite inflation, and that will cause a downturn in US treasuries. With such small spreads left, there is little scope for emerging market bonds to de-couple. With the scope for big gains largely exhausted, why take the big risks?
Nick Dewhirst is CEO of www.investors-routemap.co.uk
'Truly making a difference'
Avoidance, evasion and non-compliance
From 6 April 2019
Marcus Brookes appointed CIO