In the last few blogs, I have looked at issues impacting return expectations. In this post I go back to the Rob Carrick article and look again at the assumptions laid down by a number of commentators. These are summarised below:
If these assumptions are used to determine withdrawal rates, very few allow for the risks of a lower growth, deflationary environment (or for that matter a negative real growth/stagflation environment) and cannot incorporate the impact of multistage return scenarios on portfolio withdrawals
At the time of the article (9 June) the 10 year Canada Bond yield was 1.8% and the S&P/TSX dividend yield was 3.2%. The assumptions in the article were based on a 60 equity/40 bond split portfolio: adjusting for this we find the actual equity return component seen in the 3rd column. Expected equity returns range from 12% to 3%, with most occupying the 8 to 9% range –less dividend yields, capital returns ranged between –0.2% to 8.9%..
From 1980 until the end of 2005, long term compound total returns to Canadian equity investment were around 9.6% per annum(geometric annual), while 10 year bond returns were some 12% per annum. From 2000 to 2005, annual returns fell to some 6.6% per annum and from the end of 2005, returns have been limited to dividend yields. Since inflation has averaged some 2.3% per annum since the end of 2005, real total returns to equity investment have been negligible even before taxes, management fees and transaction costs.
Going forward it is difficult to see equity returns anywhere near historical norms of close to 10% (Century to century rates), or even within the lower bounds of the historical rolling 30 year range of between between 8% and 13%. Rolling 10 year average annual geometric returns have also, at times, been much lower than 8%:
As discussed in earlier posts, low interest rates, which many use as a reason for judging valuation, may be implying a weak expected nominal growth environment: so nominal returns anything close to very long term historical rates (+/- 2%) are at risk of not being met, by a wide margin.
The following chart shows declining historic nominal GDP growth rates for Canada, something which has been seen globally in developed economies. GDP growth globally has been slowing down of late even in the presence of low interest rates and substantial other monetary stimulus. GDP growth could slow further for reasons discussed in a number of related recent posts.
Corporate profits are above historical averages in Canada and well above levels associated with difficult economic conditions – this is again a global phenomenon. For many of the return expectations noted in the table, assuming no valuation expansion, earnings would need to expand by some 8% to 10% per annum, which does not allow for either a reversion to historical profits as a % of GDP or a risk event that causes a decline in earnings and GDP.
While Canada has benefited from rising commodity prices over the last decade (prices which have been weakening), a construction boom and a less affected banking sector, earnings growth and GDP growth from this point forward may well be exposed to a number of risks that would depress equity returns:
Excessive consumer debt, high home prices and very strong construction growth is likely to constrain activity going forward – so much latent domestic demand is being held hostage by high debt levels and one hell of a lot of the assets of the financial system are tied up in highly valued non productive assets, which as we have seen the world over, is a very big risk to the ability to finance growth going forward.
Financial sector earnings and employment could well be at a peak for a number of reasons – a detailed description is outside the scope of this article, but deleveraging, lower investment returns on assets under management and weaker consumers are all likely to limit and expose retail based earnings, while weak macro economic factors are likely to impact institutional and corporate business returns. The growth of the financial sector has gone hand in hand with financial leverage and it is likely that as we deleverage and the opportunities for returns on asset management and traditional finance remain under pressure, this sector could likewise decline in terms of economic prominence. The financial sector is a very important component of total return in the Canadian stock market.
Lower trend growth in developed economies and a medium term slowdown in China and India is likely to place pressure on commodity based earnings, production and employment.
High exchange rates are going to place continued pressure on manufacturing earnings and employment for some time. At the moment growth is therefore overly dependent on Canada’s commodity based industries and immigration.
The high nominal GDP growth of the 1960s, 70s and 80s are behind us, as is probably the higher growth from 2000 to 2007, unless we have an emerging market led boom in growth of world trade. Additionally, the impetus of declining interest rates from the early 80s onwards (increased returns on capital and allowed higher debt loads to be assumed by the economy) on economic activity is also clearly well behind us. From this point on we have a much more complex global and domestic macro economic outlook, one that would suggest that much more conservative equity return expectations need to be built into investment planning.
It is possible, in a deleveraging cycle that real GDP growth could be between 0.5% and 1.5% per annum over a full business cycle, with corporate earnings growth lagging GDP growth given that corporate profits are at the upper boundaries of their historical relationships .
Return expectations could be as low as 3.5% nominal (or less) over a full business cycle (7 to 10 years) – a negative 1.0% (real earnings growth) + 3% dividend yield + inflation 1.5% = 3.5%. Even if earnings growth moves in line with GDP, a 1% real earnings growth (if GDP grew by a real 1%) + 3% dividend yield and a 1.5% inflation would result in 5.5% nominal return. After 2% management and transaction costs (which is at the low end for many investors who are exposed to higher fees and costs) you are looking at 1.5% to 3.5% and after tax, 1.2% (40% marginal tax) and 2.8%.
Short term returns over a three to 5 year period could also be being impacted by sharp earnings/price declines. Therefore multi period returns could be as low as –5% to –10% over any given 3 year period, 3.5% to 5.5% over the business cycle (until global economies delever and rebalance) and then a higher longer term say 2.5% real earnings + 3% dividend yield + 2.5% inflation = 8% nominal, 5.5% real total return before taxes, fees and transaction costs etc. Returns could dip further if we see valuation ratios drop and, later in the cycle (10 years +) we may see P/E expansion adding to total return expectations.
All in all, if we are managing portfolios to meet liabilities we need to set conservative return and risk assumptions to minimise the risks of low return/higher risk environment. many of the return assumptions provided in the article, if they are used to based portfolio withdrawals, are not capable managing risks to return. One of the problems is of course, if you have set, in the past, higher expected returns, you are already having to pare down planned withdrawals and other investment objectives.
If you are now having to set even more conservative return assumptions in order to manage risks to return, then, you are raising questions about the efficacy of your prior planning. Many advisors with significant legacy assets structured around higher return assumptions are just not going to be able to credibly reduce return expectations to ones that are better able to manage risk. Additionally, many are operating services on cost levels that become impractical at low expected rates of return.
Anyone who uses mutual funds with deferred services charges in the range of 2.5% to 3% per annum are going to need an 8% + nominal rate of return just to earn a real dividend yield for the client assuming an average inflation rate of 2.5% from the table above. In other words, if you drop your rate of return to cover risk, you risk cutting your own throat.
One point is worth making: the return assumptions used by “advisors”/advisors are often very close to those proposed by many research documents produced by financial institutions and by academics using MPT based methodology. In other words they have not gone out on a limb. This does not mean that those who should have known better are blameless, because the risks to return should have been evident for some time.
By all means have an optimistic return expectation, but base your investment planning on tougher, meaner assumptions that a) model structural and cyclical market and economic risks and, b) have a better chance of being met in adverse conditions and c) that allow return expectations to be revised up as markets decline and ne revised down as market valuations and economic cycles mature.