Managing Risk Through Multiple Allocation Models

In “Shedding Light on the Future: Asset Allocation and Risk Management in a Post-Credit Crisis World,” researchers argue that the use of multiple allocation models is necessary to reduce blind spots and offer downside protection during “black swan” events like the 2008-09 financial crisis.

“No single model can predict the future,” says Erik Knutzen, NEPC’s chief investment officer and one of the white paper’s authors. “But applying multiple tools within a dynamic framework can facilitate better investment decisions.” 

The paper asserts that total investment program risk management is an ongoing process, requiring constant learning and the development of new allocation models. In addition, models need to be applied within a dynamic framework that can incorporate qualitative insights. 

Specific risk management strategies addressed in the white paper include mean variance, risk budgeting, scenario analysis, liquidity analysis and economic factor analysis. The paper describes each strategy in detail, and outlines how individual strategies can be merged for more effective risk management across a portfolio.

Here’s a short list of highlights from the paper:

• Investing is concerned with the future and the future is unknown. No model can change that.
• Models are lenses through which we view aspects of the future. Models, by definition, are incomplete, but they are tools which have the potential to facilitate better investment decisions.
• Portfolio theory is a useful way of thinking about future returns, diversification and risk-taking. As with any model, it has limitations. It has been subject to several misunderstandings—most notably, the idea that standard deviation is synonymous with risk—that unduly limit its acceptance.
• Too much reliance on any one model can lead to an investor being blindsided by a fat tail or black swan.
• Risk is not always measurable, but it is a useful discipline to attempt to measure it. Multiple models should be used for as much perspective as possible, though one shouldn’t take one’s own measurements too seriously.
• Risk management is the process of identifying what can go wrong with contemplated courses of action, and taking steps to keep the likelihood and magnitude of bad outcomes within tolerable limits. The more risk one takes, the more challenging risk management is. Therefore, additional models, such as scenario and liquidity analyses, may be beneficial.
• It can be very useful for investors to ask themselves, “What is the least risky thing I can do?” This question can clarify thinking and provide a useful benchmark against which additional risk and return can be measured.

The full text of the white paper is available here.