When evaluating the historical performance of an investment strategy, we often rely on a composite, which aggregates portfolios invested in that strategy behind the mask of a single group. Asset managers create composites to reassure investors that their products have a proven track record. Past success doesn’t guarantee future results, of course, but composites do tend to figure into investors’ expectations of performance and risk.
Return figures for a composite can be compelling to separately managed account (SMA) holders deciding where to invest their money. Using a composite to assess a strategy’s pretax performance is a relatively straightforward task. What matters most of all for an SMA strategy is what happens after the provider applies active tax management and other client customizations. This is where a number of other variables come into play that have a significant role in overall composite performance. Let’s look at what those variables are.
How does portfolio construction affect SMA composite performance?
A composite doesn’t necessarily include every account invested in the relevant strategy. A manager who prepares a composite has some flexibility in deciding which accounts to include. For example, we can ask these questions when reviewing the manager’s methodology:
We should also consider the risk profiles of the portfolios in the composite. Some investors may direct managers to align tightly with target benchmarks when it comes to sectors, industries, and individual securities. Others may be comfortable with higher tracking error, which means taking on more active risk. Both approaches have the goal of maximizing tax alpha. But higher allowed tracking error gives a manager room to manage taxes more aggressively, which may result in different pretax and after-tax returns between individual portfolios.
Of course, this isn’t an either-or proposition. Managers can monitor portfolios daily, take a trigger-based approach to loss harvesting, and incorporate risk factors that aim to limit the likelihood of performance deviations from the benchmark. All have the potential to provide investors with a balance of tax management and risk management.
It becomes more difficult to realize losses in an SMA over time, especially in rising markets. Individual securities in the SMA can appreciate, but its cost basis—the original value of the portfolio’s total assets—can decline at the same time due to loss-harvesting activity. The impact of both phenomena becomes apparent when we compare the tax alpha of older and younger portfolios within a composite. A composite that contains a larger percentage of newly funded accounts will appear to have higher after-tax returns than those of a composite with a greater number of aged accounts. For SMA providers just entering the space, their composites may only consist of a handful of years and limited assets. On the other hand, a more established competitor’s composite may include accounts that are a decade old or more.
One way to refresh a portfolio’s cost basis is to contribute cash. This enables the manager to purchase new positions, which in turn helps identify new opportunities to harvest tax losses. The frequency of cash contributions is another reason we see differences in composite performance. Each portfolio has a different record of net cash inflow and, with that, different after-tax performance. Even if the overall result is higher tax alpha for the composite, that figure doesn’t necessarily reflect better management.
The bottom line
An SMA’s composite can be a helpful tool, but the composite’s performance only tells part of the story. The manager’s main purpose is to build portfolios that reflect each investor’s values, goals, and preferences. This is why investors should be cautious when choosing the right SMA provider, especially when it comes to newer providers with limited track records and less experience and resources. Characteristics such as tenure, technology, scalability, expert access, tools, resources, customization, and reporting give a picture of a provider’s value far beyond return figures.