構建創造超額回報及降低風險的投資組合

Key takeaways:

• In the current higher-for-longer interest rate environment, we believe more sophisticated techniques are needed to capture alpha (excess return earned on an investment above the benchmark’s return) with better downside protection.
• In addition to bottom-up idiosyncratic security selection, we believe a thoughtful quantitative portfolio construction and risk allocation process can potentially produce consistent and uncorrelated excess returns against peer groups, which can benefit well-diversified investors.

Behavior bias and market inefficiencies

In the typical bottom-up fundamental research framework for corporate credit selection, it can be difficult for an analyst to recommend a high-quality corporate bond that trades with very tight spreads relative to its benchmark or peer group because this bond would only outperform if riskier bonds perform poorly. Furthermore, investment managers tend to be overly optimistic about their ability to forecast investment performance, which creates a bias toward riskier credit market holdings. Such behavioral bias results in potentially higher portfolio risk versus the benchmark, or beta. Another potential result is the recurrence of portfolio positions that are, in aggregate, long-carry (a tendency to try and out-yield your benchmark) and long-beta versus the benchmark, which exposes clients to excess return volatility that is dependent on the direction of markets. An example would be if a period of market distress and risk-off sentiment prevails, portfolio underperformance would be a likely outcome as the portfolio isn’t beta neutral and has more risk than its benchmark.     

Before we dive into alpha derivation from portfolio construction, we would like to point out that by hiring active managers, clients acknowledge that indexes or financial markets are inherently inefficient. The value-add of active management is that we deconstruct the credit universe into subsets to take advantage of inefficiencies and attempt to generate additional alpha, while doing a better job of incorporating our research ideas into portfolios and producing a better risk-adjusted outcome on a consistent basis.  

Bucketing the opportunity set according to volatility

Within the asset management industry, corporate credit research teams are typically structured along Bloomberg-style industrial sector classifications. While Franklin Templeton’s investment-grade (IG) corporate credit team is also organized along industry lines, we add in an extra dimension of analysis by asking our analysts to rank securities from the highest to lowest quality based on their forward-looking views and to project how spreads should trade relative to each other; the resulting conversation identifies which bonds are cheaper or riskier relative to their peers, in our view.

Not all issuers in the same industrial sector are similar in terms of credit ratings or other idiosyncratic factors, thus, risk profiles differ from issuer to issuer. For example, a high-quality A rated technology issuer typically has tight long-duration bond spreads relative to the index, whereas a BBB rated technology issuer typically has wider spreads versus the index and is very volatile. Thus, high-quality versus low-quality technology companies have little in common with each other in terms of risk and volatility.

In order to better bucket risk across a portfolio, we tri-furcate the sector universe into low-beta, mid-beta and high-beta buckets based on historical spread volatility. Low beta can be thought of as the highest quality corporates (such as issuers that have the tightest spreads relative to the index with the lowest volatility).  An example of mid-beta would be higher-quality and less cyclical BBB rated credits, and high-beta is typically the more cyclical lower-quality BBB rated credits, such as BBB commodity issuers. We think this approach enables a better portfolio construction process, not only from a bottom-up perspective, but also by isolating sectors and separating the riskier portions from the higher-quality portions for better risk allocation. As we rank high-quality to low-quality issuers, the takeaway is there is a need to bucket beta risk more effectively so that the opportunity set is truly competing with the more appropriate peer group.

Strategic placement of opportunity set along the yield curve

Credit managers have generally sought to increase their investment returns by taking on more duration risk than the benchmark. While this strategy has typically worked to investors’ advantage when rates stayed low and spreads were contained, it also exposed investors to possible losses if interest rates or market conditions moved against their positions. As we enter a higher-for-longer interest-rate regime with forecasts of greater volatility in risk assets, we believe the longer duration and higher beta risk formula for adding returns should be reexamined. Interest-rate movements are inherently unpredictable, in our view. Accordingly, we have adopted a duration-neutral approach as we see duration management as a less consistent source of alpha relative to idea generation and portfolio construction.

As the credit curve reflects the yield levels at different maturities, we instead identify what we believe to be the most efficient place to own risk by taking positions at the steep points on the curve that will provide additional return from roll-down (a technique where an investor would own a bond and hold it for a period of time so that the value of the bond increases as it moves closer to maturity as it is valued at a lower yield). In instances where credit curves are flat—or even inverted—there is no additional spread compensation for owning risk beyond a certain maturity, and we would avoid doing so. In short, placement of bond positions along the credit curve can be extremely additive to portfolio returns given the different shapes of the credit curve across sectors and ratings.

In determining the maximal roll-down positioning on the credit curve, we also consider where the optimal carry point is on the credit curve (or the point where the return from holding a bond is optimal). By grouping credits together based on their risk profile (as alluded to earlier), we are able to better evaluate the relative carry potential within these various volatility categories.

Putting it all together

We believe our unique portfolio construction method allows us to incorporate our best idiosyncratic security selection ideas, while enabling beta neutrality relative to our benchmark, through risk bucketing according to volatility tiers. A benefit of beta neutrality is market directionality becomes less of a factor in alpha generation, especially in the current market environment where it’s hard to ascertain whether markets will respond positively to good news.

Our portfolio optimization process also strives for duration neutrality relative to our benchmark by using strategic positioning along the credit yield curve. We believe the resulting portfolio isolates and takes advantage of a benchmark’s structural inefficiencies while attempting to offer downside protection from adverse market conditions.

Our portfolio construction framework can be easily adapted to fit the needs of a diverse spectrum of investors, from buy-and-hold insurance portfolios to active strategies with higher turnovers that take advantage of investment opportunities as they present themselves. We believe we have a portfolio construction process that is repeatable with consistently positive outcomes that are less correlated with our peers. In our view, clients need diversification in their managers as much as they need diversification in their fixed income portfolios.