CPD: Modern Portfolio Theory and the Rise of Ratios in Portfolio Construction

Background

Since their inception, financial markets have provided investors with the opportunity to increase their wealth. However, this is of course not always the case – at times, and without warning, these markets see dramatic price declines or volatility, resulting in the significant decrease in investors’ wealth. These moves may be the result of a weakening economy, a systemic shock to a particular market, or the ending of a period of irrational enthusiasm by investors. Regardless of the driving force of these moves, investors must remain vigilant and well positioned to ensure that these periods of volatility do not cause permanent damage to their investment portfolio.

During the latter half of the 20^th century, significant advancements were made in the field of financial economics. These developments provide a superior framework for investors to manage and understand portfolio volatility and risks. One of the most significant developments was Harry Markowitz’s concept of Modern Portfolio Theory (MPT). A vital element of the theory was the formalising of an asset’s risk as the standard deviation of its returns, and in turn the formation of an efficient frontier of possible investments. This acknowledgement enabled investors to compare the risk-return profile of two assets and select the one with the highest return for a given risk, or the lowest risk for a targeted return. Another significant outcome of MPT was the recognition of how investors could achieve the optimal level of diversification within their portfolios.

While investors recognised the value of diversification prior to MPT, it was more of an art than a science, with successful portfolio managers relying on asset-picking and market timing. Both activities have proven all but impossible to implement successfully on a consistent basis across market cycles. Other investment strategies involved simple rules-of-thumb, such as 1/3 equities, 1/3 bonds and 1/3 real estate, or simply aiming to maximize returns.

Post MPT, new heuristics developed around the role of each asset class. Equities became known as the potentially higher returning assets, but with these returns came higher risk. Alternatively, the fixed income sector was seen as a source of lower but more stable returns. Therefore, an aggressive (conservative) investor would maintain a portfolio of 80% (60%) equities and 20% (40%) fixed income. Crucially, this high-level definition of fixed income investments did not discriminate between the various sub-classes within the fixed income universe. One class which is discussed later is credit assets.

The development of a formal asset pricing model was the next stage of evolution within the field of financial economics. The initial model, the Capital Asset Pricing Model (CAPM), divided risk into systemic risk (risk that cannot be avoided via diversification) and those risks particular to a given asset. From the CAPM came two significant metrics, the Sharpe ratio, and the information ratio (IR). The Sharpe ratio, developed by one of the originators of the CAPM model and Nobel Memorial Prize in Economic Sciences winner William Sharpe, became the accepted term for measuring risk-adjusted returns.

Equation 1 defines the Sharpe ratio, with:

representing an assets or portfolio’s return minus the risk-free rate, defined as its excess return. To identify the excess return per unit of risk, the excess return is divided by the portfolio’s volatility, as measured by its standard deviation.

With the general acceptance of the Sharpe Ratio, the desire for additional metrics grew. One such example was the need for a metric to compare the variations of returns across, and within, asset classes.  The IR was adapted to assess these nuances. The key difference is that the risk-free rate is no longer included, replaced by the relevant benchmark for the portfolio. For instance, an Australian large cap manager would utilize the ASX200 as the benchmark, while a domestic credit manager may use a composite bond index.

While the Sharpe and Information Ratios have been widely adopted, they are not without fault. As seen in Equation 1, the denominator is the variance of a portfolio’s return, where these returns are assumed to be normally distributed. When returns display asymmetric results, the Sharpe ratio loses some of its relevance.

One point of difference between credit and equity securities is the risk of default. In this instance, default refers to an issuer of a given security not paying the legal coupon and/or returning the entire principal at maturity. In such an instance, the security is at a heightened risk of losing a considerable amount of its value. In comparison, a company is never obliged to continue dividend payments, nor to return any capital as shares exist in perpetuity. Therefore, a credit manager must manage market risk – that is the movement in credit spreads – and credit risk, which is the risk of default. In comparison, equity managers will look to manage the market risk of their securities as prices move in response to new public information.

Regarding the IR, the selection of a relevant benchmark is crucial. This choice is relatively straightforward for equity managers, but for fixed income managers the availability of an investable benchmark can be problematic. For example, a credit manager may invest across a broad range of securities, for example corporate bonds, private credit, structured assets, or distressed debt. All these assets have very different characteristics, meaning it is difficult to find a single relevant benchmark. Regardless of these issues, both ratios can make a meaningful contribution to the process of portfolio construction.

Australian case study

This section provides a brief practical example of how one might look to utilise MPT and associated metrics to construct an Australian domiciled investment portfolio. There will be a particular focus on the implications for assessing and selecting credit funds. The data also provided insights into the appropriateness of the two metrics across the various asset classes.

The figures and data used in this exercise are the 3 year (annualised) returns, associated standard deviation and relevant ratios sourced from the Morningstar Direct database for Australian diversified credit, and blended Australian mid/small and large capitalisation funds. Critically, this analysis is based solely on past returns and does not forecast future returns.

Figure 1 places the return profile of funds within the three fund categories in the classic risk/return space (returns on the Y-axis and risk/standard deviation on the X-axis). In general, the results are consistent with expectations that equity funds, on average, provide higher returns, but these come with greater risk (higher standard deviations). Additionally, within the equity space small/ mid-cap funds are riskier, with a large return variation within the peer group. Over the same period, credit funds delivered in general lower returns with lower risk, again consistent with the belief that credit funds will deliver relatively stable returns, at the cost of an uncapped upside.

Having identified an appropriate asset allocation, the next question is how to select the appropriate fund(s) for one’s portfolio. This analysis will mainly focus on the diversified credit segment. Prior to assessing the available credit funds, it is crucial to define the expectations around what one can expect from a credit fund. In general, a credit fund’s returns will predominately come from the yield of the securities within the portfolio. For an Australian fund, this yield will be greater than the official cash rate as set by the Reserve Bank of Australia (RBA).

The gap, known as the spread, primarily depends on the risk profile of the securities within a fund. The other source of return will be capital returns. This return fluctuates as the market price of credit securities fluctuates in a manner consistent with other risk assets. This capital return will also be affected if a particular security defaults or faces a ratings downgrade.

Returning to the expectations regarding a credit fund, it is most likely that investors will be after stable returns to offset the volatility stemming from the equity component of their portfolio. Therefore, the Sharpe ratio and IR are well placed to provide meaningful insights. From Figure 1, it is evident that there is a noticeable variation in risk and return for the 3-year numbers. Importantly, given the generally lower returns of credit funds, the effects of these variations become evident through an assessment of the Sharpe Ratios across the investment universe.

Figure 2 illustrates the variation in Sharpe Ratios across the three asset classes used in this paper. The first point that becomes apparent is that despite the variations in return and risk, the equity funds are quite tightly bunched. This position contrasts with the figures for the credit segment, where not only is there a broader spread but a noticeable proportion of the population returning a negative Sharpe ratio. The ramification of this characteristic is that managers were unable to outperform the cash rate, an outcome which is a red-flag when selecting any fund.

What are the implications of these findings? Credit funds offer compelling risk adjusted returns in comparison to equities. More importantly, the evidence suggests that investors need to carefully assess those funds to ensure they meet their investment objectives, which is primarily to provide a stable income flow with limited downside risk to the capital value of the investment.

What are the likely characteristics of a credit manager who can meet these requirements? At a minimum they are likely to be able to adjust the following attributes of their portfolio to meet the prevailing market conditions:

• Duration: A portfolio’s duration reflects how much the capital value of the portfolio will vary with a change in credit spreads. If spreads are expected to tighten (loosen) then a higher (lower) duration is appropriate. Therefore, a credit manager can underperform if they have extended their duration in the hope of improved market conditions, only for these conditions not to eventuate.
• Credit Risk: A credit manager can increase (decrease) their yield by going further down (up) the capital stack, as determined by a security’s credit rating. If a manager is anticipating benign conditions and increases their credit exposure, returns can be diminished because an issuer defaults on their payments or low rated securities de-rate further in a risk-off environment.
  
  

Figure 2 provides the data on the IR across the three selected asset classes. In contrast to the findings of assessing the Sharpe ratio, the results are far more informative for the equity funds than the credit funds. Informative in the sense that there is a greater variation in the results, which in turn enables investors to identify those funds which have outperformed on a risk adjusted basis. The data also provides an insight into the shortcomings of IR for assessing credit funds, that is the median IR is materially higher than those of the equity funds.

The origins of the contrasting IR results are seen in Equation 2. Utilising a benchmark as opposed to the risk-free rate allows one to identify the better performers within an asset class. This characteristic is particularly useful for equity funds where there is an obvious and, more importantly, investable benchmark. Indeed, from Figure 3 one can see that over 3 years the median large cap manager failed to better their benchmark, identifiable via a negative IR, indicating that an index fund may be a more appropriate investment. Alternatively, in the small/mid cap space the median manager has added value and there are several managers that have performed well.

Regarding the credit fund universe, there is not the dispersion of IRs within the sample. A partial explanation is that there simply is not the same quantum of opportunities to add (or destroy) excess value via a small number of positions that vary greatly from the benchmark – a common strategy for equity managers.  Compounding this point is the difficulty in establishing an effective benchmark because the credit markets are wide, with some sections lacking depth. However, the more relevant point relates to the purpose of a credit fund, which is to deliver lower but stable returns. Therefore, to identify the credit managers that have performed well the Sharpe ratio is more appropriate.

Conclusion

The advent of MPT brought a great deal of science and formal theory to portfolio construction. These developments allow investors to construct portfolios that meet their individual needs. However, there certainly is not a “one size fits all” approach, and it’s important to understand which ratio or approach is most appropriate for assessing their investment options. MPT has also allowed investors to clearly identify the role of each asset class within their portfolio and provides the ability to assess the risk-return characteristics of each class. As we’ve seen, credit funds were assessed as appropriate for investors seeking stable returns. This fact is not to say that excessive returns are not available in credit markets, rather to say that these sorts of returns are the exception. Another quandary for credit investors is to find an appropriate benchmark which offers a meaningful hurdle for managers to better. Therefore, assuming that past performance is no guarantee of future performance, the Sharpe ratio is the most appropriate way to compare the risk-adjusted returns of credit funds.