Acadian’s MACS strategy employs a systematic investment process that generates recommended portfolio allocations on a daily basis. This systematic process is made up of four key steps, which taken together translate fundamental data into tradable portfolios:
- Factor-Based Return Forecasts
- Adaptive Risk Model
- Portfolio Construction
Our views are expressed via return forecasts that we apply to all assets of the MACS universe. To obtain these return forecasts, the MACS investment team has developed a broad set of factors which evaluate assets from different perspectives. In selecting these factors, achieving diversification across factor forecasts and understanding how factors interact with each other are key considerations. Factors fall into two basic categories: the first category captures asset-specific characteristics, associated with themes like value, carry, momentum and quality; the second category considers the impact of exogenous factors on asset prices, which includes macroeconomic themes such as growth, inflation, and stimulus. The individual factor forecasts are combined to generate the aggregate return forecast for each asset. The forecasts adapt on a daily basis to changes in the markets, and consistent with our defensive bias, more weight is put on factors that perform well in down markets, such as quality-type factors.
Moving from a set of return forecasts to a robust portfolio requires an understanding of the risks associated with the underlying assets, and of the correlations across those risks. For a well-constructed portfolio, returns associated with individual portfolio positions should balance out with the risk they contribute to the portfolio. Thus, even if return forecasts were to remain unchanged, changes in the risk environment should lead to changes in portfolio positioning. Acadian’s MACS strategy uses a proprietary risk forecasting tool, which dynamically adjusts risk forecasts with changes in the risk environment. In order to capture these changes and to avoid noisy forecasts, the model incorporates different horizon risk measures and blends them in an effort to provide the most predictive forecast.
Portfolio construction starts with return and risk forecasts, in conjunction with applicable constraints and objectives, to arrive at a mix of exposures seeking to maximize return and minimize uncompensated risk. Moreover, in an effort to protect the portfolio on the downside, the portfolio construction process considers two types of risk, which are handled in different ways:
- Foreseeable risks are risks that we understand and that can be captured by our risk model. As such, they are addressed via our portfolio construction process, which rewards return but imposes penalties on total portfolio and asset class level risk, risk concentration, outsized relative value positions, etc.
- Unforeseeable risk is associated with difficult-to-forecast events like natural disasters. In an effort to protect against such risks, hard limits are imposed on position sizes and asset class exposures, floors are established on (over-optimistic) volatility forecasts, and greater uncertainty is assigned to smaller assets.
A typical portfolio is implemented primarily using liquid derivatives to efficiently gain the desired exposures from the construction process and minimize transaction costs. To build a portfolio of tradable instruments, the asset exposures from the portfolio construction step are translated to tradable instruments. A specific mapping system aims to match asset exposures and tradable securities in a way that minimizes the basis risk between the two and while reducing trading costs. The portfolio is then traded in the most cost-efficient manner.
Implementation will typically involve shorting and the use of leverage, which are a means to achieve diversified and well-balanced portfolios. When building a portfolio, we think about its underlying positions in and the risk they contribute to the overall portfolio. Once the portfolio construction process gives us the desired risk level for a given position, we then seek to determine the notional exposure required to achieve that risk. We expect gross leverage, which is the absolute sum of these notional exposures, to average approximately six times portfolio NAV. Because of the implicit leverage provided by the use of derivatives, this approach tends to be cash efficient, leaving the larger part of the portfolio in unencumbered cash.