Date of Award
Summer 8-22-2026
Document Type
Dissertation
Degree Name
Ph.D. in Business
Organizational Unit
Daniels College of Business
First Advisor
Conrad Ciccotell
Second Advisor
Jack Strauss
Third Advisor
Tracy Xu
Copyright Statement / License for Reuse
All Rights Reserved.
Keywords
Decumulation, Long-term care, Monte Carlo, Portfolio solvency, Retirement spending, Sequence of returns risk
Abstract
Retirement income planning is shaped by a foundational tension between the fear of portfolio depletion which drives systematic conservatism, against the unidirectionally of time making foregone early-retirement consumption an equally consequential planning failure. Conventional planning is anchored in static withdrawal assumptions with fixed models which fail to reflect the nonlinear reality of a typical retirement lifecycle. This dissertation shows that for a substantial percentage of retirees, the balance between what their portfolios can sustainably support in early years of spending is meaningful even when considered with the potential constraint of self-supporting the cost of probability weighted long term care.
The dissertation comprises two papers. Paper One investigates whether retirement portfolios can sustain front-loaded distributions during the active "go-go" phase of early retirement. Specifically, an additional withdraw of 20% of the portfolio value across the first two to ten years without compromising long-term solvency. Using parametric Monte Carlo simulations, historical back-testing, worst-case stress testing, and bootstrap resampling applied to a 65% equity/35% fixed income portfolio, results demonstrate that all front-loaded scenarios exceeded the 80% minimum solvency threshold at the 20- and 25-year horizons. These horizons are consistent with actuarial data placing the typical retirement duration for a 65-year-old at approximately 19.7 to 20.7 years. The 30-year horizon produced failure across all scenarios, confirming the compounding sensitivity of extended decumulation periods to elevated early withdrawals.
Paper Two extends this framework by incorporating back-end demand of longterm care costs across five care scenarios, as well as front-loaded spending considerations. This study reviewed the probabilities of paid care outcomes and durations. Home health or assisted living of one to two years, affects approximately 32% of retirees and reduced the 30-year portfolio success rates by only four to six percentage points from the baseline of no long-term care. Front-loaded early retirement spending remains compatible with self-funding anticipated care costs for most retirees.
Collectively, these findings support a shift away from static withdrawal frameworks and toward evidence-based, phase-specific distribution planning that honors the temporal structure of retirement, the actuarial realities of longevity, and the compounding cost of underconsumption during the years when utility is greatest.
Copyright Date
8-2026
Publication Statement
Copyright is held by the author. User is responsible for all copyright compliance.
Rights Holder
Eric Michaud
Provenance
Received from ProQuest
File Format
application/pdf
Language
English (eng)
Extent
132 pgs
File Size
3.1 MB
Recommended Citation
Michaud, Eric, "An Examination of the U-Shaped Nonlinearity in Retirement Spending" (2026). Electronic Theses and Dissertations. 2789.
https://digitalcommons.du.edu/etd/2789