Joe Fernandes, head of the global investment solutions group at CFSGAM, says the firm does not believe in tactical asset allocation but rather has a “reward – for – risk philosophy” for its multi-asset portfolios. As a portfolio manager, he says the levers available within TAA among a small number of asset classes or sectors are limited relative to stock selection across a much broader universe of opportunities. “For our institutional investment solutions business, our starting point is to understand the investment objectives of our investors,” he says. “We build portfolios aligned to those objectives or to the investor’s liability profile. For those portfolios, the starting point isn’t the SAA at all, it’s the liability profile of the investor. We then look amongst all the sources of risk that are available for those that will contribute to the achievement of the investment objectives. The asset allocation decision is a derivative of the risk allocation decision.” He says the concept of DAA is intuitively appealing, particularly following the extreme market events of last year, but the challenge is in the execution. “It’s possible to be right in your investment view but for markets to move against you. Getting the timing right is a real challenge,” he says.
BTIM does not describe its allocation process as DAA, but uses a combination of valuation-based longterm asset class forecasts, with frequent reviews and tilts if necessary, and a TAA overlay through the Global Macro group which is designed to make money from high frequency trading. “You’ve got three things to worry about: one is the return, one is the volatility of the return, and the other is more esoteric – it’s the way in which asset prices tend to move together, or the covariance of the return,” Swift says. “We are happy to set a policy mix but let the exposures ‘drift’ with the market. This is a way to let the portfolio benefit from the tendency for asset class returns to show serial correlation – that is they can go on long upward or downward moves. “In setting any policy mix you need to acknowledge that you’re going to be wrong. Be humble in your forecast, there’s a very good chance that you’ll be wrong in one if not all of those three key variables. When you add the complexity of the risk taken by the investment process within the asset class you have lots that can go wrong.