The Right Portfolio Construction Questions

Summary: Fund managers can benefit from consistently asking the right questions in their portfolio construction process. A coherent understanding of both how the investment ideas are generated and the desired shape of the portfolio results in a more investable final investment product. Making all the inputs in the portfolio construction process explicit can improve manager decision making and open the door for using systematic portfolio construction tools to build better portfolios.

In an investment fund’s portfolio construction process, front office teams generate investment ideas and create an investable portfolio for their investors.

Here we’ll discuss a framework for thinking about the fund management portfolio construction process.  This framework helps fund managers consistently answer the core questions required to build portfolios that deliver the best expression of their investment ideas.

Fund managers can improve their decision making and build better portfolios by following this type of process-driven approach to portfolio construction.

Portfolio Construction Process Overview

The portfolio construction process in fund management consists of two steps:

  1. Generate Investment Ideas (ie. Generate Alpha)
  2. Apply the Ideas to Create an Investable Product (ie. Shape the Portfolio)

Portfolio Process Illustration

An ideal portfolio construction process translates the investment ideas generated by the front office team into an investable product that best expresses the Alpha in the ideas, whilst complying with the objectives and constraints of all portfolio stakeholders.

An investable portfolio meets investor requirements and is consistent with the product and sales messaging of the fund management organization.  When the fund management team is able to consistently produce and deliver Alpha in an investable portfolio all stakeholders benefit.

The training and background of most fund managers largely focuses on Step 1 – generating ideas.  Analyst training, CFA certifications, an MBA and more provide tools for generating ideas.  From the very beginning and throughout their career managers develop their ability to generate and evaluate investment ideas.

However, constructing the best portfolio to deliver the alpha in their ideas (Step 2) poses a very different and complex problem. Yet doing so in a consistent manner is key to delivering the portfolio results desired by the fund’s LPs and growing their business.

Regardless of the specific methods ultimately used to solve the portfolio construction problem and build the portfolio, fund managers can benefit from a structured process to understand and codify the inputs.

This article, and others in this series, will discuss some of the frameworks Sherpa helps managers implement to better express and retain their Alpha in their portfolio construction.

Process Step 1 – Generate Alpha

There are many different approaches to generating Alpha.  Many managers believe their edge comes from a superior understanding of company fundamentals.  For others it may be the result of quantitative analysis. Some rely on technical analysis, or some blend of different disciplines.

Whatever the USP of a given team, the first step in the portfolio construction process requires understanding how the investment ideas are generated.  In short – how does the team generate their alpha?

Two questions to consider:

  • How do you determine which assets to include in the portfolio? (Selection)
  • How do you differentiate between the assets’ prospects? (Scoring)

For the purposes of setting up the portfolio construction problem we refer to step 1 as creating an Alpha Vector.  The Alpha Vector is a list of the assets to include in the portfolio with relevant scores and attributes that differentiate them in building the portfolio.

How Do You Determine Which Assets to Include?

Understanding how investment ideas are added to the portfolio plays a fundamental role in setting up the portfolio construction problem. Defining the Selection process serves as a critical starting point for portfolio construction.

Benchmarked geographic or sector equity portfolios require a different risk framework than single stocks in a wider multi-asset book.  Deeply researched long-term fundamental value plays imply a different set of constraints than higher turnover tactical ideas.  How often do ideas come in and out of the portfolio?

How Do You Differentiate Between the Assets’ Prospects?

Once the manager has selected the assets for inclusion in the portfolio, the next step is to establish a rubric for expressing forward-looking views for each asset.

Scoring the assets lets the fund manager differentiate between various ideas on conviction, quality, expected return or any other metric.

A conviction or alpha score can be as simple as a binary high-low conviction or a 1-5 ratings system.  Other teams base their scoring around projected values such as traditional asset price target or a forward-looking view on expected returns.

The process of differentiating between assets is embedded in most managers’ selection process.  Establishing a forward-looking view on the asset is critical to determining if it should be included in the portfolio.

Others might have benchmark or legacy positions included in their portfolio without developing a clear view on the asset.

Some managers may choose not to differentiate between their assets, effectively assigning them all an equal “score” in the portfolio construction process. This decision is no less viable than establishing a more developed scoring system, but should be understood as a clear decision while setting up the portfolio.

Process Step 2 – Shape the Portfolio

Once the assets have been selected and scored, the next step is to shape the portfolio.  Shaping the portfolio involves building a complete set of all portfolio objectives and constraints from all portfolio stakeholders to ensure the portfolio is investable for the end LPs.

Answering these portfolio shape questions often involves inputs from multiple stakeholders in the organization to determine what portfolio shape best fits investor preferences and is thus most investable.

  • Which risk(s) do you want to take? (Risk You Want)
  • Which risks do you NOT want to take? (Risk You Don’t)
  • What additional operational constraints are on the portfolio? (Objectives and Constraints)

What Risk Do You Want to Take?

The first step in shaping any portfolio is being explicit in defining the Risk You Want.  This must be consistent with what the fund’s ultimate investors want from the fund and how the fund is marketed and sold.  For the resulting portfolio to be investable it should take the risks investors want, while mitigating those they don’t.

The fund’s mandate is one key element to understand here. For instance:

  • A long-only expression of the investment ideas will retain more market exposure
  • A long-hedged expression may remove exposure to the broad market but retain residual sector, factor or other exposures in the ideas
  • A long-short expression of the ideas may seek to isolate and maximize the specific risk

In addition to looking at what types of risk are inherent in the mandate for the fund, the manager should also determine which Scoring component(s) within the Alpha Vector they want the most exposure to.

As an example, say the assets have been scored on Expected Return, an ESG score and the coverage analyst’s conviction.  How should exposure to these different scores be prioritized?  Should risk be driven primarily by higher Expected Return ideas, or are all three scores of equal importance?

Explicitly defining the risk you want ensures that the portfolio is being built to both express the alpha within the team’s investment ideas while delivering the risk exposure promised to investors.

What Risks Do You NOT Want to Take?

Once the manager has defined the risks you want to drive the portfolio, the next step is to determine which Risks You DON’T want the portfolio to take.  This sets up the portfolio to take the desired LEVELS of risk while mitigating unintended, residual sources of P&L movement.

Different LPs may have very different preferences about the amount and types of risk acceptable in their portfolio, leading to either compromises (or separately managed accounts with different risk profiles).

In this process stakeholders need to determine risk limitations including:

  • Total level of risk, such as a volatility or drawdown threshold
  • Relative level of risk, such as tracking error versus the benchmark or market beta
  • Residual risks, such as momentum, growth/value, or any other critical factor

What Additional Objectives and Constraints Are on the Portfolio?

After specifying the Risk You Want versus the Risk You Don’t, the final step in setting the portfolio construction problem is defining relevant stakeholder objectives and constraints.

A complete view of the objectives and constraints is critical in ensuring the resulting portfolio is realistic, implementable and fulfills the requirements of all portfolio stakeholders.  Setting up the appropriate constraints helps ensure that the ultimate portfolio is investable.

For instance, a portfolio created entirely of small-cap stocks might not have enough liquidity to capture the alpha in the names at the scale required by an institutional client.

When thinking about portfolio objectives and constraints it can be useful to consider them at multiple different levels:

  • Asset Level constraints apply to individual assets in the portfolio, and could include liquidity, turnover, size, or any other factor or attribute at the individual asset level.
  • Group Level constraints apply to groupings of assets within the portfolio, such as sector groups, sub-strategies, or any other way the portfolio is organized.
  • Portfolio Level constraints define characteristics of the full portfolio – simple examples such as the number of stocks or net/gross exposure, or complex examples such as beta to a macro risk factor, ESG factors or any other attribute.

Use the Answers to Help Build the Portfolio

This basic framework helps fund managers ask the right questions to ensure their portfolio construction inputs are well-defined and understood.

Ask yourself these questions when building your portfolio:

Generate Alpha
    1. How do you determine which assets to include in the portfolio?
    2. How do you differentiate between the assets?
Shape the Portfolio
      1. Which risk(s) do you want to take?
      2. Which risks do you NOT want to take?
      3. What additional operational constraints are on the portfolio?

The process of defining these inputs shines a light on the portfolio construction process and gives fund managers the data they need to improve their decision making and build portfolios that deliver the best expression of their alpha to their investors.

Many fund managers focus on only the first part of this process and leave too many of the inputs ill-defined. Then the final step of building the portfolio has been done in the portfolio manager’s head. As a result the mathematical process and resulting portfolio is what we might call “fuzzy” at best.

However, having a full understanding of all inputs in the portfolio construction problem opens up a range of more systematic options for those who wish to utilize them. These systematic portfolio construction methodologies can help managers build better portfolios and deliver results for their investors.

Future posts will discuss some of these methodologies in more detail.

For more on Sherpa’s Process Alpha approach to portfolio construction, check our Process Alpha Archive here.

To see the potential impact of improved portfolio construction on your portfolio and learn more about refining your team’s portfolio construction process, get in touch with the Sherpa team or sign up for the SFT mailing list here.