Global bond markets are changing structurally. Those changes receive much attention, and rightfully so. However, we should be careful not to lose track of the forest while minutely examining the trees. For most investors, the prospect for bond market returns – income plus capital gains – is far more important than the prospective changes in the composition of the markets. In this regard, investors are likely to find that past returns are a poor predictor of the future.
The coming five years promise an evolution in the composition of global bond markets, and perhaps a revolution in investors’ expectations about prospective returns to bonds (and equities, too).
As recently as one year ago, conventional wisdom was that the US Treasury market was going to dry up and disappear. This would profoundly change the world’s bond markets, because the US Treasury market is the most important, if no longer the largest fixed income market in the world. (That honour belongs to the Japanese government bond market. In fact, the US Treasury market is also smaller than the US mortgage-backed securities market.) The US Treasury market’s importance derives from its role as a benchmark against which many other securities are quoted and priced, and because its bell weather status means that changes in the US Treasury yields usually cause sympathetic changes in the rest of the world’s fixed income markets.
The US Treasury market has received a stay of execution. The market was expected to shrink because the US government was expected to enjoy budget surpluses, “as far as the eye can see,” as one US government official characterised it. It turns out that that particular official suffered from myopia. The US budget surplus was in part an artefact of the long US expansion and the bubble economy of the 1990s. As a result of the 2001 recession and prospective slower growth, as well as expected spending increases by the US Federal Government following the terrorist attack of 11 September, the US is now expected to record a budget deficit for at least each of the next three years.
Although the US Treasury market is no longer shrinking absolutely, it is expected to shrink relative to GDP. So, too, government bond markets in Europe. That is because in both the US and Europe budget deficits are not growing as quickly as the economy. Not so in Japan, which is expected to run large government deficits, “as far as the eye can see”. As a consequence, the Japanese share of the global government bond market - at 28%, now greater than that of the US (25%) – will grow even larger.
So one thing to watch for in coming years is that Japan will claim an increasing share of global government bond-only indices. That is one reason many investors are abandoning government-only benchmarks, in favour of more comprehensive indices containing asset-backed securities and corporate bonds. Since the markets for ‘spread products’ (non-government debt) are small in Japan, using broader benchmarks helps mitigate the ‘Japan problem’ – disproportionate representation of Japan in global bond indices.
A second major change in the global bond market will come in Europe. In Europe the corporate bond market, and the markets for asset-backed securities, have been growing rapidly since 1999, and that trend will continue. This trend is the result of both supply and demand forces. The supply of corporate bonds and asset-backed securities results from the attempt by European banks – facing a more competitive commercial environment in Europe following EMU – to increase return on equity by reducing the size of their balance sheets. As banks sell off assets they create asset-backed securities for bond investors, and when banks reduce their willingness to make loans they force their corporate clients to issue bonds instead. The demand for ‘spread paper’ in Europe is the result of a sharp reduction in government bond yields in many countries, such as Italy, where bonds used to trade at a considerable yield spread compared to Germany. The compression of government yield spreads in Europe is also a result of EMU.
Most investors want to know what returns they can expect from bonds during the coming five years, more than they want to know how the market’s structure will be changing. Of course, we cannot be nearly as certain about prospective return as we can about structure. However, we can draw some broad conclusions. The most important is that recent history may be a poor guide to bond returns during the coming five years.
Figure 1 shows that US bond market total returns have declined from 10% per annum at the end of the 1980s and early 1990s, to about 7% in the second half of the 1990s. World government bond returns have also averaged 7– 10% for much of the 1990s.

These are attractive total returns, but can they be repeated? Most investors base their asset allocation decisions on forecasts of long-run returns that are based on examining a long history of historic returns. Implicitly they are assuming that the future will look like the past. Unfortunately, three not-to-be-repeated events have dominated bond returns during recent history.
First, there was a global disinflation. Inflation and inflation expectations – it is the later that is key for bond yields – fell sharply during the 1980s and 1990s. That lowered yields, and produced capital gains for bondholders. But now inflation is low, so disinflation is unlikely to occur again, unless inflation first increases.
Second, the Japanese economy slipped into chronic recession during the 1990s, as a result of the collapse of the stock market bubble of the late 1980s, and of poor economic policies. The result was deflation in Japan, and extraordinarily low bond yields.
Third, EMU meant sharply lower inflation and bond yields in many European countries.
All these one-off events produced lower inflation, and falling bond yields. That meant capital gains for bondholders. So bondholders have come to expect capital gains that top up the income component of their return. However, falling yields and the resulting capital gains bond investors earned historically cannot be duplicated going forward, unless the entire globe slips into a Japanese-style deflation. Very simply, the US and Europe already have achieved inflation of 2–3%; there is not much further for inflation to fall.
Suppose that going forward, inflation (and interest rates) in the US and Europe no longer fall. Then the total return enjoyed by government bond holders will be no more than their coupon income – about 5%. In Japan, unless deflation intensifies, investors will be lucky to earn even the coupon which, at a paltry 1.3% on a 10-year JGB, seemingly can only rise from here. So we are forced to the uncomfortable conclusion that the total return to global government bonds during the coming five years is likely to be 5% or less, not the 7–10% that many investors have grown accustomed to. As figure 1 illustrates, 5% total return is very much on the bottom end of historic experience.
That makes it sound like bond prospects are unattractive, but consider the alternatives. Bonds may look pretty good compared to equities. US equities are expensive, priced at very high price-to-earnings ratios, and at very low dividend-to-price yields, compared to history. As a result, P/E ratios are unlikely to grow from here. That means that US equity returns will be no more than earnings growth. In the long run, corporate earnings growth is governed by the growth in nominal GDP. Since GDP is not likely to grow faster than 6–7% in our current, low inflation world, equity investors should prepare for a new world of lower returns, too.
So, are prospective bond returns high or low? Both. In absolute terms, prospective return of 5% or less looks low. But a prospective return which is only 1% or 2% below that of equities seems historically generous for bonds, relatively speaking.
Lee Thomas is chief global strategist of PIMCO, based in Newport Beach, CA