Investment ManagementRisk ManagementTechnology

Data-Driven Decisions Aided by Dashboards and Scorecards

By November 5, 2020 No Comments

A NeuGroup member describes his money market fund dashboard and investment manager scorecards.

Dynamic dashboards that help corporates leverage data to make better decisions are becoming essential tools for finance teams committed to tapping technology to transform.

  • Some companies in the NeuGroup Network are generating envy by using in-house tech talent skilled in programs such as Python to create dashboards, while other members are turning to Power BI or Tableau to ramp up.
  • But a recent meeting of treasury investment managers underscored that it’s the data in a dashboard and the decisions it prompts that matter most, whether the dashboard showcases liquidity, cash flow, ESG ratings or—in this case—money market funds (MMFs).

An MMF dashboard. One member’s MMF dashboard intrigued peers, who asked not about its whiz-bang technology (it’s compiled in Excel using mostly ICD data) but how it’s set up and how often the company refreshes the data in it.

  • The company’s treasury team designed the dashboard internally about five years ago and included metrics to help assess any vulnerability in funds given the then-pending impact of reforms involving gates and fees.
  • “It provided some early warning signals that in one case allowed us to exit a fund which ultimately folded,” the treasury investment manager said.
  • These days, the dashboard is updated every two weeks, except when there is market stress or other reasons to review funds more closely. It’s used by the member to monitor risk and positioning and by team members who make buy/sell fund decisions.
  • “It serves as an early warning system for any fund-related issues, which allows us to proactively position ourselves and optimize risk/reward,” he said.

Facts and figures. The member’s dashboard contains about 25 MMFs, mostly prime and some government, both US and offshore. In the future, he said, it may be automated using RPA to save time and perhaps “allow us to add more variables without manual work.” It currently shows:

  1. Balances
  2. Yields
  3. NAV volatility
  4. Fund AUM/trending
  5. Fund concentration
  6. Key stats, e.g., 7-day liquidity

Fund manager scorecards. The same member described to peers another way his team makes use of data generated on spreadsheets—fund manager scorecards that are turned into PDFs.

  • Among other criteria, managers are rated on trade settlement, compliance and the value they add through research, events and idea generation. And, of course, performance:
  • “We try to look at sources of performance and see how that might translate into our policy guidelines,” the member said. “For instance, a manager who relies heavily on derivatives might not perform as well when the tool is not available to them.”  Other key factors:
    • Annualized returns vs. a benchmark (net of fees)
    • Tracking error/difference vs. benchmark
    • Volatility
    • Qualitative; five P’s: people, philosophy, process, performance, price

Value added. “The real benefit I see from manager scorecards is that they ensure a consistent and structured two-way conversation with our managers,” the member said.

  • “It surfaces issues for discussion and everyone knows where they stand.  So when we are adding or, unfortunately, subtracting assets, it’s typically not a surprise because they’ve seen some consistency of feedback.”
0
0
Justin Jones

Author Justin Jones

More posts by Justin Jones