Episode 393: Engineering Financial Outcomes

Episode Summary

Executive Summary: The episode argues that financial planning should be treated like engineering: start with real client goals, define constraints, then use models and sensitivity analysis to quantify trade-offs. Brayden Warwick shows how software like Conquest and PWL’s free retirement tools can reveal material differences between strategies such as salary vs. dividends and CPP/OAS timing—while also warning that planners must understand model limitations and not rely on generic scorecards.

Main Topics: Engineering mindset applied to financial planning (Priority: 5/5): Brayden frames planning as an optimization problem: identify objectives, constraints, and trade-offs rather than treating projections as the plan itself. Goals first, not surface-level metrics (Priority: 5/5): The discussion stresses that clients often say they want higher returns or lower taxes, but planners must uncover deeper life goals before modeling anything. Sensitivity analysis and trade-off quantification (Priority: 5/5): The episode emphasizes finding which variables move outcomes most and using charts like Pareto frontiers to compare competing goals such as spending vs. retirement age. Software power and limitations (Priority: 4/5): Conquest and other tools are presented as powerful, but only if planners understand how outputs are calculated and what they omit; otherwise, scores can be misleading. Case study: salary vs. dividends for incorporated business owners (Priority: 5/5): A Canadian business-owner example shows that the default retirement score can mislead, while adding the correct legacy goal dramatically changes the optimal choice. Case study: CPP/OAS deferral and sustainable spending (Priority: 4/5): The free retirement tool is used to show how timing CPP/OAS affects sustainable spending, and how planning for bad outcomes can support more confident spending decisions.

Key Arguments: A financial projection is not the financial plan; projections are inputs used to make decisions, just as engineering models inform designs. Good planning starts with the client’s true goals, which are often different from their first stated goals like maximizing returns or minimizing taxes. Planning tools should be used to evaluate outcomes against the correct objective; default software KPIs may be irrelevant or misleading if not matched to the client’s goal. Sensitivity analysis helps planners focus on the variables that actually matter most for a specific client scenario. When clients have multiple goals, planners should quantify trade-offs rather than dismissing the math because it does not capture every real-world nuance. Pareto frontiers are a practical way to show multiple optimal solutions and help clients choose among them based on preferences. In the incorporated-business case, paying salary improved the retirement score but paying dividends massively improved legacy value, proving that the “best” choice depends on the goal. CPP and OAS deferral should be evaluated through present value and real planning outcomes, not simplistic break-even-age logic. A planner must understand the mechanics and assumptions under the hood of software outputs to avoid making recommendations based on misunderstood metrics. The profession should communicate uncertainty professionally and explicitly, similar to medicine or engineering, to build trust and improve outcomes.

Data Points: Podcast episode: 393 - Episode number for the discussion on engineering financial outcomes. Webinar date: February 12 - Live webinar on how much you need to retire in Canada. Webinar time: noon Eastern Time - Scheduled live webinar with Q&A moderated by PWL financial planners. Conquest retirement score with salary: 150%+ - Base case in the business-owner example when paying salary from the corporation. Conquest retirement score with dividends: 135% - Alternative case in the business-owner example after removing salary. Legacy value with salary: $16.5 million - After adding a legacy goal to the salary scenario in Conquest. Legacy value with dividends: $34.1 million - After adding a legacy goal to the dividends scenario in Conquest. Legacy value difference: about $18 million - Difference between dividend and salary cases for the incorporated business owner. Spending under base case: $53,600 per year - Sustainable spending level in the retirement tool for the bad outcome assumption. Spending under base case: $4,467 per month - Monthly equivalent of the sustainable spending level in the retirement tool. Adjusted spending with OAS at 70: $53,800 per year - Small increase in sustainable spending when OAS is deferred from 65 to 70. Spending increase from OAS deferral: $200 per year - Incremental gain in sustainable spending from deferring OAS in the example. Bad outcome viability: ~84% - One standard deviation below the mean, used to approximate plan viability for the retirement tool. Terrible outcome viability: ~97.5% - Two standard deviations below the mean, used to approximate a more conservative viability threshold. CPP/OAS spending drop after depletion: about $10,000 per year - When assets run out in the example, spending falls from about $54,500 to about $44,889. Retirement age knee point example: age 62 - Illustrative Pareto frontier knee point where spending increases more sharply for one extra year of work. CPP age in example: 70 then 60 - The case study compares deferring CPP to 70 versus taking it at 60 to show impact on spending and portfolio longevity.

Pivotal Quotes: "the only thing that we know for certain about this plan is that it's wrong" — Ben Felix / discussed by Brayden: Used as the common but problematic way planners describe uncertainty, which Brayden argues against. "a financial plan is equivalent to a design for an engineer" — Brayden Warwick: Core analogy for treating planning as an objective-driven design problem rather than a prediction. "the projection is not the plan" — Brayden Warwick: Key distinction between modeling outputs and the actual advice or decision being made.

Implications: Planners should stop optimizing for generic software scores and instead model the client’s actual goals, constraints, and trade-offs. Better software literacy and sensitivity analysis can materially improve advice, trust, and outcomes.