While almost everyone in the investment industry agrees with the fact that asset allocation is a key driver of overall portfolio performance, they are divided as to how to do it, what the appropriate asset classes are, whether asset class weights should be dynamic or fixed and how asset allocation should be determined.
Modern Portfolio Theory argues the case for strategic or fixed asset allocation based on investors’ risk profiles and performance objectives. There is also a lot of support for a tactical or dynamic asset allocation strategy – which is typically the result of a top-down investment process.
However, Linda Eedes, senior analyst at RE:CM, says that a more reliable method of asset allocation is as a by-product of a bottom-up, valuation-driven process which focuses on buying cheap securities on a case-by-case basis, and defaulting to cash in the absence of sufficient opportunities. Eedes says that it is also possible to allocate capital effectively amongst asset classes using this approach, providing an additional source of return to bottom-up stock selection.
She says there are a number of problems with strategic asset allocation as advocated by Modern Portfolio Theory. These include using volatility as a measure of risk, diversification benefits which often aren’t there when you need them most, valuations not being taken into account and reliance on historical inputs.
An alternative to strategic asset allocation is dynamic or tactical asset allocation, which moves among various asset classes to take advantage of short and intermediate market inefficiencies as a source of return. Many investors use a top-down process to determine which asset class to allocate capital to, which relies heavily on forecasting factors such as the economy, interest rates, sectors that will do well within that environment and stocks that will do well within that sector.
Eedes, however, says forecasting is a dangerous game. “Let us assume you are correct 70% of the time, which in itself would make you a statistical anomaly. If you require all forecasts to be correct in order for your investment thesis to work, you have only a 24% chance of getting it right. Given all the other forecast inputs a comprehensive analyst’s model requires, this number in reality is probably even smaller.”
She adds that consensus forecasting has completely failed to predict any of the last four recessions. In addition, studies show that analysts forecasting companies’ earnings two years prior to the actual event are wrong by a staggering 94%. Even at a 12-month time horizon, they are wrong by around 45%. Finally, in 2008, analysts in the US forecasted a 24% price increase, yet stocks eventually fell by 40%.
Eedes says that the good news is that once you reject forecasting as a waste of time, it frees you up to focus on the things that a disciplined investor can measure with a reasonable degree of accuracy. “We agree with Milton Keynes when he said: ‘I’d prefer to be approximately right rather than precisely wrong.”
She says a better alternative is a bottom-up investment philosophy based on thorough, fundamental research, which assesses each investment on a case-by-case basis. Using this approach, investors aim to buy undervalued assets, sell overvalued assets and wait until the market corrects these perceived misvaluations. Eedes notes that this is very different to market timing, where investors try to call the top and bottom of the market. “Decisions on whether to be invested in or out of the market, or bonds or any other asset class is an outcome or by-product of a bottom-up process. If you can’t find sufficient ideas in equities, property or bonds to meet your criteria, it is better to default to cash.”
She says the starting point when looking for investment ideas is the equity market. “We know this is the best generator of returns over the long-term. The biggest consideration when limiting the risk of capital loss is to ensure that you do not pay more for an asset than what it is worth. When it comes to returns, you should be concerned with real capital growth in absolute returns rather than nominal relative returns versus a market index. You should look to pay for sustainable earnings levels rather than peak earnings, and look for them to be protected by either the value or the underlying net assets of sustainable companies with competitive advantages such as barriers to entry and pricing power.”
She says when it comes to bonds, these need to offer a yield with adequate inflation protection over time. “If you believe that bond yields do not offer this compensation, it is better to defer to cash.”
Eedes says that evidence that a bottom-up approach to asset allocation works is the fact that the RE:CM Global Flexible Balanced Fund, has outperformed the All Asset Allocation Unit Trusts Universe by 3% annually since its inception in 2003, delivering annual returns of 10% over inflation on an annualised basis after fees.
“Investment managers need to be able to show that whether their process is top-down, bottom-up driven or a combination thereof, it has resulted in effective asset allocation over the full cycle, protecting capital by avoiding overvalued assets and growing capital by taking advantage of significantly undervalued assets. The extent to which valuations are cheap or expensive at the start of an investment horizon has a significant impact on returns thereafter. Consequently the idea of valuation-indifferent fixed-weight asset allocation strategies simply makes no economic sense.
“Investors who have a time horizon of at least five years should always focus on generating the best long-term real returns while limiting the risk of capital loss,” concludes Eedes.
Get the latest issue of FAnews