The relationship between risk and return is seldom static for long, but most strategic asset allocation processes pretend that it is. Tactical asset allocation can help address this issue, says Sandhurst Trustees’ Thad McCrindle.
When institutional investors set a strategic asset allocation (SAA), they need to make a number of assumptions about the direction of capital markets in the coming years.
Often these capital market assumptions are expressed as return expectations for the various asset classes over a 10-year period, presented in bands depending on the level of conviction investors have in the forecast returns.
The problem with this model is that risk and return are relatively static throughout time, but in reality the relationship between risk and return can vary quite dramatically.
To deal with short-term dislocations, investors can implement tactical positions that skew a portfolio towards a certain outcome.
Thad McCrindle, Chief Investment Officer at Sandhurst Trustees, implemented a tactical asset allocation (TAA) program shortly after the Global Financial Crisis (GFC) and has been fine-tuning the model ever since. Over the past 10 years, he estimates it has added about 50 basis points to the portfolio annually.
“Risk-return opportunities are not consistent through time; they definitely change. There are situations and conditions where the chances of a really bad outcome are quite low and vice versa,” McCrindle says.
Risk-return opportunities are not consistent through time; they definitely change. There are situations and conditions where the chances of a really bad outcome are quite low and vice versa
“I am talking about equities when I say that, but there’s always the equivalent in other asset classes like bonds, credit and other liquid asset classes.”
TAA addresses non-stationary risks in the market, he says, and he likes to think about these risks in terms of probabilities.
“What I mean by non-stationary risk is simply that there are conditions over shorter periods of time, whether that’s a few months or a year or a few years, where the risk and return profiles do diverge from your SAA,” he says.
“We want to adjust our asset allocation to be aware of those things over that shorter timeframe.”
He gives an example of a recent market event that warranted a closer look.
On 5 August 2024, the Japanese stock market crashed and the Nikkei 225 index fell 12.4 per cent in a single day. But the reasons behind the crash seemed technical, rather than based on a deterioration of company fundamentals.
“We were having a conversation in early August, where we as a team could clearly agree that mean reversion was going to occur. And the only question simply was: ‘Where’s the trigger to generate that mean reversion?’” he says.
“It was entirely evident that there was forced selling occurring in both the Nikkei and in the unwinding of a short yen position, so forced buying of yen in essence.
“You can see people clearly haemorrhaging a position and once that ends it goes back the other way. So that’s where we were elevating some of those signals that have a mean reversion component to them, noting that it will probably be really short term.”
McCrindle and his team were proven right as only a week later the Japanese market had recovered nearly all of its losses.
“There are moments where you can see market behaviour is much more likely to mean revert and that’s kind of where it’s good to have a portfolio of signals,” he says.
The Model
The TAA model McCrindle has developed is a combination of fundamental models and quantitative signals, but essentially it centres on three pillars: momentum, valuation and sentiment.
“A lot of people will probably have momentum and valuation in their process, but the other one is positioning, sometimes we use the term ‘sentiment’. It’s really about the information you can aggregate to estimate where the faster money is positioned,” he says.
“Since March 2020, we have observed that positioning seems to be getting extreme quite often. And really when I say positioning, we are thinking about when something is crowded, because then we get cautious.”
Since March 2020, we have observed that positioning seems to be getting extreme quite often. And really when I say positioning, we are thinking about when something is crowded, because then we get cautious
McCrindle and his team use a series of different timeframes and can take positions based on their outlook for the next month up to the next three years. But the way they implement their trades is dependent on the interplay between the different signals.
For example, they don’t just consider a momentum signal on its own. It needs to be viewed in the context of the current level of pricing in the market and the level of crowdedness of a trade.
“When you are valuation aware it constrains you after your momentum has been working for a while and prices get higher. It doesn’t constrain you typically when it starts working,” McCrindle says.
“The other one is sentiment; when positions become quite crowded that actually makes us more sensitive [to cut a position].
“We will implement and set our positions discretionarily. So we will move when something is on the expensive side and perhaps approaches a valuation that is significantly away from average.
“We don’t care if it is a little bit expensive, but when something is expensive and the positioning is tight, then we will be taking profit.”
COVID
Momentum can be a fickle strategy. Although it provided a decent level of protection during the GFC, when markets tumbled 40 per cent and more, the performance of momentum strategy since has been somewhat erratic.
For example, during the March 2020 drawdown on the back of the COVID-19 pandemic, momentum didn’t have enough time to set in as equity markets rebounded sharply only a few weeks later.
McCrindle, who describes his approach toward momentum as being on the slower side, rather than trading short-term signals, acknowledges momentum was of little use during the pandemic.
“Momentum didn’t help that much, for sure, but it did wake us up to stop waiting for another leg down. It was a stop loss on bearishness, if I can put it that way,” he says.
We did a decent-sized switch from nominal bonds to inflationary bonds. The market was pricing as if inflation was done. It was going to be zero forever. We figured that it probably wasn't going to be negative and so it was a very asymmetric opportunity in our minds
Yet the valuation part of the TAA process identified some interesting opportunities during this period.
“We did a decent-sized switch from nominal bonds to inflationary bonds. The market was pricing as if inflation was done. It was going to be zero forever,” McCrindle says.
“We figured that it probably wasn’t going to be negative and so it was a very asymmetric opportunity in our minds. That was really by far our most successful trade, certainly in the last five years or so. By the end of 2020, linkers had outperformed nominal bonds by nearly 20 per cent.
“Which is enormous in the context of that part of your balance sheet, being the defensive side of your balance sheet.”
Thad McCrindle spoke at the [i3] Asset Allocation Forum 2024, which was held on 22 & 23 August in Terrigal.
__________
[i3] Insights is the official educational bulletin of the Investment Innovation Institute [i3]. It covers major trends and innovations in institutional investing, providing independent and thought-provoking content about pension funds, insurance companies and sovereign wealth funds across the globe.