Making sense of the macroeconomic environment is complex. Figuring out implications on instruments is an even more daunting endeavor.
And yet, markets often rhyme; patterns repeat themselves. Is there a system which can exploit historical similarities and remove most of the forecasting complexities for the end user?
We have approached this question by introducing a systematic macro regime framework, called Pharos, which explores recurring patterns in history.
Based on similar periods in the past, future behaviour can be extrapolated and used in asset allocation, equity factor tilts, or even to drive client conversations.
We infer macro regimes through various tradable instruments that capture distinct dimensions of the macro complex, such as VIX levels, interest rate spreads and equity prices.
We only take the most representative instruments for each asset class, avoiding redundancy. When multiple instruments are needed, we orthogonalize them to extract their unique contribution to that macro dimension.
This setup is able to represent the macro environment of any given day.
For instance, economic risk will be reflected by equity volatility, and copper is widely seen as a leader for the economy.
The Pharos framework considers macro regimes as fluid concepts over time. It calculates similarity in a continuous manner as the "distance" between the observed macro factors.
The benefit of fluid regimes over the common discrete methodologies is that more granular comparisons are possible between periods.
In a 3-state regime framework, one can only infer that two periods are in the same or in different regimes. With a fluid methodology, one can go further and infer the exact (dis)similarity of two periods.
The end of 2022 provides an interesting case study. The economy was at a crossroads, with inflation proving hard to contain. The market was in a fragile state and many expected an imminent recession.
What could be seen in Pharos was the following:
1. Equities were trending down,
2. FX was holding strong,
3. VIX was elevated but stabilizing,
4. Commodities were trending lower,
5. Yield spreads were increasing and
6. Interest rates were rising.
The Pharos framework then sets out to find periods in time with similar conditions.
These are the aftermath of the internet bubble at the start of the century and the height of the Great Financial Crash in '08. Other similar periods coincide with periods of stress such as the COVID-19 crash.
The most dissimilar periods were predominantly periods where crashes were ending and very calm periods (such as 2017).
As market patterns tend to repeat over time, we expect returns in subsequent periods to show a high degree if similarity as well.
Pharos is therefore able to infer forward returns for any set of instruments.
Pharos creates an aggregate estimate of the forward returns based on the behaviour of past coincident regimes. We show these estimates for the end of 2022 together with the observed outcomes in the graph below.
All of the projected returns are directionally correct.
Momentum and Minimum Volatility especially end up being correctly seen as large laggards.
Pharos is therefore an excellent companion for tactical positioning within markets.
In this case study, we use Pharos estimates to drive equity style allocations. We consider Value, Momentum, Growth, Low Volatility, Quality and Size.
The static portfolio has an equal weight to all styles, while the dynamic portfolio uses projected returns to set the allocation mix. These weights are smoothed out over three months. Short positions are allowed.
The resulting portfolio values are shown on the graph below.
Because of the ability to short styles when projected returns are negative, draw-downs are smaller than the static blend (-25% vs -55%).
The dynamic allocation also allows for faster recoveries coming out of bear markets. This leads to an annual performance of 14% against 9.3% for the static blend, a significant difference.
Similar to the previous case study, we can use Pharos estimates to drive asset class allocations. We consider a very simple universe containing equities, gold and treasuries. The static portfolio has an equal weight to all three asset classes, while the dynamic portfolio uses projected returns to set the allocation mix. These weights are smoothed out over three months. This strategy is long-only.The resulting portfolio values are shown on the graph below.
Even within this very simple setup, Pharos is able to profitably navigate the different regimes.Annual returns are 8.8% versus 7.2% for the fixed blend. Draw-downs are similar.
Making sense of the macroeconomic environment is complex.
Figuring out implications on instruments is an even more daunting endeavor.
Pharos, our systematic macro regime framework, explores recurring patterns in history. Based on similar periods in the past, expected returns for instruments and asset classes can be extrapolated.
Our use cases showed the power of Pharos to dynamically allocate to equity styles and asset classes.
It is also an excellent quantitative tool to assist in client communication. It provides an easy to understand framework to set client expectations for future returns.
Informed clients are more at ease, leading to less stressful conversations when things get rough.
Pharos. Your guidance partner in markets.
Pharos assists wealth managers in setting their allocation mixes. It is also a powerful quantitative tool to assist in client communications.
