Week 12 Day 2: Bad Years Early Can Be Fatal

In year 2, you are withdrawing from a larger base. A 30% crash wipes out $300,000 from a $1,000,000 portfolio. You still need to withdraw $40,000. You are now drawing from $660,000 instead of $960,000. By year 15, you have already spent down some capital, so the base is smaller and you have fewer years of withdrawals remaining. Early crashes hurt most because they compress the base when it matters most.

Let us run the full math. $1,000,000 portfolio, $40,000/year withdrawal, 7% average return. Scenario A: -30% in year 1, then 7% every year after. Year 1: $1M becomes $700K minus $40K = $660K. Year 2: $660K grows 7% = $706K minus $40K = $666K. Notice you are barely treading water. Year 5: approximately $705K. Year 10: approximately $897K. Year 30: approximately $1,680K. Survived, but barely. Scenario B: 7% every year, then -30% in year 15. Year 15 balance before crash: approximately $1,800K. After crash: $1,260K minus $40K = $1,220K. Year 20: approximately $1,460K. Year 30: approximately $2,500K. The early-crash scenario left $820K LESS at year 30, despite identical total returns. The difference: $1,680K vs $2,500K. One bad year at the start cost $820,000 over 30 years.

The mathematical explanation for SOR risk lies in the interaction between multiplicative returns and additive withdrawals. Portfolio evolution with withdrawals is: V(t+1) = V(t) * (1 + r(t)) - W, where W is the withdrawal amount. This is a linear recurrence relation where the multiplicative term (1 + r) and the additive term (-W) interact asymmetrically: when V is large, the multiplicative return dominates. When V is small (after early losses), the additive withdrawal becomes proportionally larger and more destructive. This is why Bengen found that the worst historical retirement periods were not those with the worst average returns but those with the worst early returns. The 1966-1995 cohort (worst historical SWR) faced a combination of the 1973-74 bear market AND high inflation, which devastated both real portfolio values and real withdrawal amounts simultaneously. The compound annual return of that period was actually decent -- but the sequence destroyed early retirees.