A Reliable Framework Helps Manage Risk

The term “risk management” is universally espoused in the investment community but not necessarily universally implemented. At Brandywine Global, the ultimate purpose of risk management is to minimize the permanent loss of client capital. Furthermore, risk management is especially important when managing across multiple sectors that contain numerous factors driving performance. But how, exactly, do we aim to accomplish that within our multi-sector and credit-oriented strategies? There are many approaches to risk management, varying in method, complexity, and output. However, all share the common goal of making the portfolio manager keenly aware of the potential sources of future volatility. Whether it be a value-at-risk analysis, a Monte Carlo simulation, or a key rate duration report, each of these elements has an important role in the investment process.

Understanding the Limitations of Risk Management

With all this information available, what is a portfolio manager to do? We think the first step is to acknowledge and accept some of the weaknesses inherent in the majority of risk management tools. Most traditional methods use historical observances to forecast future outcomes. While this approach is sound in theory, a simple review of 2020 would suggest expecting the unexpected because “unprecedented times” may be only a pandemic away. Relationships break down due to secular shifts, and securities that once offered a counterbalance may now trade in the same trajectory. In a similar vein, we think it is important to know what is not captured as an input to some of these tools. How does a risk management system model a $7T Federal Reserve balance sheet? What about record federal deficits and seemingly bottomless monetary support? Ultimately, the data out is only as good as the data in, and if the tools being used to inform investment decisions are incomplete, the investment manager must account for and adapt to these known flaws.

Responding Predictably to Future Unknowns

Our viewpoint has always been that pure output from risk management analysis is much less important than its interpretation. We think one of the most critical steps to managing risks lies in constructing a process around managing the unknowns, that is assigning probabilities to forecasted outcomes, acknowledging incomplete information, and building a comprehensive framework of what future events may do to asset prices. This framework is at the foundation of sizing positions, assessing conviction levels, and providing true diversification across the portfolio. The process was recently put to the test surrounding the recent U.S. election. Understanding assets would perform differently under the multitude of political outcomes, portfolios were positioned accordingly based on the firm’s assessment of probabilities.

Ultimately, our goal is to use all tools at our disposal to make ourselves as informed as possible to arrive at educated probabilities of future outcomes. This process is questioned, challenged, and refined on a daily basis given adapting viewpoints from all members of the investment team. The result is a portfolio in which we feel confident to respond with predictability as future unknowns reveal themselves.

Differentiating Volatility from Outcomes

Lastly, risk management is only as good as the outcome. Managing ex-post or realized volatility is more important at Brandywine Global than managing to forecasted or ex-ante risk. In fact, in many cases, our view has been that ex-ante risk increases usually correlate to a possible opportunity rather than signaling an inherent problem. As evidenced in the two charts below, the most recent spikes in implied volatility measured by the MOVE Index were subsequently followed by significant spread compression in high yield corporate bonds.

In the end, it is not the complexity of the tools themselves or the sophistication of the risk management framework that ultimately matters. Realized drawdowns and Sharpe ratios are what define the effectiveness of risk management for an investment manager.