Research: 10 year return forecasts
AMP Asset Management head of investment strategy Paul Dyer takes a "big picture" look at what has driven recent investment returns and what returns investors can reasonably expect over the next decade.
Thursday, December 17th 1998, 12:00AM
Markets went through a clear inflection point in the early 1980s, as global inflation and interest rates began to fall sharply. We suspect they are going through another such process at present.
A Golden 15 years for Investors
Investment returns since the early 1980s have been among the highest recorded for any 15 year period. Global equities have returned a compound 14.9 per cent annually since 1983, and global bonds almost 10 per cent pa. New Zealand assets have also performed strongly. Fixed interest has returned 12.9 per cent pa and equities 8 per cent pa, despite falling 65 per cent between 1987-91. Inflation has averaged just 3.6 per cent in the US and 4.4 per cent in New Zealand over the same period.
Three key factors have underpinned these returns:
- A lengthy period of prosperity. There has been only one major global recession during this 15 year period (1991/92). Profits have grown steadily, with earnings per share growth for global shares averaging 9 per cent pa.
- Falling inflation. World inflation has fallen from an average of 7-8 per cent in the early 1980s to 1-2 per cent today. Bond yields have fallen almost continuously as a result.
- Multiple expansion. Greater comfort with the environment has seen both real interest rates decline and equity multiples expand. In the US equity prices have risen 585 per cent since 1983, but earnings have risen only 222 per cent. With constant multiples these growth rates must converge.
The last point has been especially significant. What investors need to appreciate is that they have been receiving one-time gains. When the earnings yield on an equity falls from, say, 10 per cent to 5 per cent, prices double and investors receive a huge windfall. But returns thereafter are much lower - dividend yields have halved and the scope for future capital gains is reduced. This is essentially what has happened between 1983 and 1998. In 1983 the average price to earnings ratio for the US equity market was 12x. Today it is 26x.
Obviously such multiple expansion cannot continue indefinitely.
What Should we Expect in Future?
We make longer term investment return projections for equity returns via three approaches. Each serves as a basic cross-check on the other.
Method 1: Risk Free Returns + Risk Premium
Firstly, to project equity returns, we can look at risk-free returns and likely risk premium. Here the news is not encouraging for investors. Bonds have been falling almost continuously since 1983, with notable interruptions in 1986-90 and 1994.
At present, the average 10 year Government bond for the five largest economies is just 3.7 per cent. The return generated from these bond markets over the next 10 years will be around this level. Returns from global markets have been very high in recent years as yields have been falling, generating capital gains. In addition, high short term interest rates in New Zealand have boosted returns when these assets were hedged back to NZ$. This also no longer applies. New Zealand short term interest rates are now close to average world levels. In our judgement a return to rates well above world levels is unlikely in the foreseeable future.
It is very difficult to see sustained capital gains from these levels. Indeed, a major bond market selloff is possible at some point. The current average yield of 3.7 per cent is probably an upper bound on returns that we can reasonably expect.
The equity risk premium also appears to have contracted over the past decade. One measure of this can be calculated by comparing earnings yields, bond yields and consensus inflation expectations. This produces the following graph.
Despite the rise associated with the very recent market selloff, the global equity risk premium remains only about 3 per cent. If we combine this level with the expected risk free return of 3.7 per cent, we obtain a point estimate of just 6.7 per cent for future equity returns. Even when expressing this as a range to allow for uncertainty in the methodology, we would probably not project returns of more than 6-9 per cent pa.
Method 2: Historical Real Returns
A second check is to compare prospective returns against those prevailing in the past. As noted above the past 15 years have been unusually strong. What does a longer term picture show?
The graph below shows cumulative real returns through investing in the US sharemarket during this century. A dollar invested in 1900 would be worth just over $1000 today after inflation, but ignoring taxes. This equates to a compound real return of between 6-7 per cent pa.
Note that this history has been far from smooth. Lengthy periods of stagnation were experienced between 1910-1920, in the 1930s and again in the 1970s. Each was associated with a major world shock (World War 1, great depression, oil shocks). By contrast, periods of temporary over-valuation (eg 1987) did not leave a noticeable dent in long term returns. Since major shocks are unforecastable in advance, our best guess is that expected real returns will continue at around long term averages for the next decade.
Currently world inflation is around 1.5 per cent. Adding this to expected real returns of 6-7 per cent implies nominal returns of 7-9 per cent, a similar result to that of Method 1.
Method 3: Expected Dividend plus Capital Returns
Equity returns are simply the sum of dividends received plus capital gains or losses. Currently the MSCI dividend yield is just below 2 per cent. In the long run profit growth should equate with GDP growth. Long run world real GDP growth is around 3 per cent. With world inflation around 2 per cent, growth in nominal GDP growth and hence profits of around 5-6 per cent is a sensible expectation.
Some adjustments are needed to equate this to capital returns for shares. Companies do not pay out all their earnings, and retained earnings also add to capital growth. This alone should add around 2.5 per centpa at current payout levels. Secondly, listed companies tend to be concentrated in faster growing sectors of the economy - slower growth sectors (eg Government and agriculture) are under-represented. Against this some growth is funded by new capital raising, so earnings per share growth will be a little lower. Combining these various influences, long run earnings per share growth of 6-8 per cent might be expected.
Adding the dividend yield to likely capital returns suggests a total return of 7-10 per cent. Once again we arrive at a very similar conclusion to the previous methods.
What About New Zealand Assets?
Our judgement is that over the long run, NZ$ returns will be marginally higher than international returns. The expected margins are around 1 per cent for equities and perhaps 0.5-1 per cent for bonds.
Real interest rates in New Zealand are clearly above world levels. Our analysis points to an average margin of around 1.5 per cent at the long duration (say 10 years) in recent years. Where NZ$ short term interest rates are also higher, and foreign assets are hedged back to local currency, this margin diminishes sharply.
Equity returns are also likely to be higher for local investors, especially when dividend imputation credits are included. Imputation currently adds almost 2 per cent to the market’s yield. However, the true benefit to local investors is probably less than this amount. To the extent that imputation raises prices for NZ shares the net dividend yield will be a little smaller than it would otherwise have been. Our calculations suggest that the true margin over global shares is probably around half of this, i.e. around 1 per cent pa.
Currency movements will also be influential. The NZ$ has been highly volatile in the past, and may remain so into the future. However, looking ahead there is little reason to expect either sustained appreciation or depreciation. The NZ$ currently looks close to longer run averages. More importantly, the RBNZ’s inflation target is very close to prevailing world inflation levels. This suggests that the central expectation should be for currency stability.
Projected Returns
Point estimates of future returns are obviously subject to wide margins of error. What is clear is that the very strong returns of recent times are unlikely to be repeated. Our "best guesses" for returns over the next decade for each major asset class, and for a typical medium risk balanced fund (50-60 per cent growth assets) are shown below:
The margin of error around these estimates is large. However, the message that returns will be significantly lower than in recent experience is clear.
These Returns Are Still Attractive
It is a mistake to think of low nominal returns as unattractive. With prospective inflation of 2 per cent or below, the implied real returns are around longer term historic experience.
The argument becomes even stronger when tax is considered. Investors should logically be concerned with real, after tax returns. The table below calculates these for a range of nominal returns and inflation rates, assuming a 33 per cent tax rate. Only at very low inflation rates and/or very high nominal returns are real, after tax returns positive.
Real, After-Tax Returns
Our future estimate of 6-8 per cent pre-tax returns for balanced funds equates to around 2-3 per cent after tax and inflation (which we assume to be 2 per cent). By contrast, with inflation at 8 per cent (roughly the New Zealand average through the 1980s), nominal returns in the order of 15-20 per cent would have been required to achieve these levels. With a nominal tax system, inflation punishes investors particularly harshly. Lower nominal returns, when associated with low inflation, are paradoxically much better for investors.
Conclusion
Our perception is that many clients continue to anticipate rather higher returns. In the near term they may well be right. But the arithmetic for longer term returns, based on today’s prices and yields plus likely economic growth rates, is very clear. The past decade or so has been a lengthy transition phase from the high inflation / high yield phase of the 1970s and early 1980s. Capital returns have been very strong through this period. As this phase ends single digit expected returns should become the norm into the future.
Paul Dyer is head of Investment Strategy at AMP Asset Management.
| « Portfolio Talk: David McClatchy | King builds an empire » |
Special Offers
Commenting is closed
|
|
Printable version |
|
Email to a friend |
