Multi-objective Optimization in Planning

In this episode, we are joined by Michael Kothakota for a deeply technical and thought-provoking conversation on interdependent integrative financial planning theory. Drawing from his background in academic research and real-world advisory practice, Michael introduces a mathematical framework designed to capture the full complexity of financial planning—where decisions across domains like taxes, investments, and estate planning are interconnected and constantly evolving.

We explore why traditional economic models fall short in capturing the individualized and multi-dimensional nature of financial planning, and how Michael's approach uses tools like multi-objective optimization and dynamic programming to better reflect reality. He explains how client preferences, time-varying priorities, and uncertainty all interact within the model—and why even identical financial situations can lead to very different optimal decisions. This episode is a deep dive into the mechanics of financial advice, offering a new lens on how planners can create value by integrating decisions across domains and aligning them with what clients truly care about.

Key Points From This Episode:

(0:04:00) Introduction to the episode and why this topic leans heavily into financial planning complexity.

(1:04:00) The core takeaway: integrating all financial planning domains leads to better outcomes than siloed advice.

(5:35:00) What interdependent integrative financial planning theory is—and why interdependencies matter.

(7:16:00) Why traditional economic theories like portfolio optimization and consumption smoothing fall short.

(9:37:00) The central insight: financial planning must account for structure, preferences, and time.

(12:12:00) Modeling financial planning as a complex, preference-weighted system over time.

(14:25:00) Why identical financial situations can still lead to different optimal advice.

(17:50:00) Multi-objective optimization and the competing goals within financial planning.

(21:09:00) The role of dynamic programming in solving sequential financial decisions.

(23:42:00) Evidence on whether financial planners improve client outcomes—and the limitations of existing data.

(26:58:00) The architecture of the model: structural tensor, priority weights, and discount matrix.

(30:31:00) Why financial planning is "non-smooth" and filled with constraints and trade-offs.

(33:57:00) How changing strategies over time are captured through evolving "strategy spaces."

(36:50:00) The six financial planning domains and their respective objective functions.

(42:35:00) The priority matrix: quantifying what clients actually care about.

(44:41:00) Discount rates and urgency—how priorities shift over time and with life events.

(47:58:00) Why financial planning must account for uncertainty and changing preferences.

(49:53:00) The role of financial planners in shaping and educating client priorities.

(51:07:00) The four-tier architecture that combines structure, preferences, and urgency.

(52:47:00) Capturing uncertainty: endogenous vs. exogenous risks and planning for shocks.

(55:39:00) Theoretical results: integration premium and value loss from misaligned advice.

(58:09:00) Practical takeaway: always consider cross-domain effects when giving advice.

(1:02:24) Real-world example of value destruction from siloed expert advice.

(1:06:34) Why the value of integration scales with complexity—not just wealth.

(1:07:42) The enduring importance of human financial planners in navigating complexity.

Links:

Meet with PWL Capital: https://calendly.com/d/3vm-t2j-h3p

Rational Reminder on iTunes — https://itunes.apple.com/ca/podcast/the-rational-reminder-podcast/id1426530582. Rational Reminder on Instagram — https://www.instagram.com/rationalreminder/

Rational Reminder on YouTube — https://www.youtube.com/channel/ Benjamin Felix — https://pwlcapital.com/our-team/

Benjamin on X — https://x.com/benjaminwfelix

Benjamin on LinkedIn — https://www.linkedin.com/in/benjaminwfelix/

Editing and post-production work for this episode was provided by The Podcast Consultant (https://thepodcastconsultant.com)