There's a detailed explanation of the calculations of new worksheet in this post of the VPW forward test thread.Monel wrote: ↑Fri Oct 25, 2019 11:44 amUsing the simpler approach, the real life dedicated bridge funding is subject to stock market variability and should be included in assessing downside risk. Can you confirm that using VPW to assess the risk associated with a 50% decline in equities that VPW correctly subjects the bridge funding to the risk of loss?longinvest wrote: ↑Thu Oct 24, 2019 4:52 pm
There are two approaches to bridge Social Security. There's the older approach of using a dedicated CD/TIPS ladder or savings account as explained in the post Delay Social Security to age 70 and Spend more money at 62. That's what you illustrated in your example, above, with a dedicated $200,000 bridge.
A simpler approach is to use the new VPW Accumulation And Retirement Worksheet with a single retirement portfolio. The investor could, for example, put the entire $1,000,000 into a 40/60 stocks/bonds portfolio, like the Vanguard LifeStrategy Conservative Growth Fund (VSCGX), and simply use the worksheet to determine withdrawals. This newer approach is illustrated in details, creating monthly retirement income, in the thread A Simple Bogleheads Retirement Using Variable Percentage Withdrawals (VPW Forward Test).
The old and new approaches differ; it shouldn't be surprising if they result into slightly different outcomes.Monel wrote: ↑Fri Oct 25, 2019 11:44 amAm I correct that VPW now applies the loss to the entire portfolio before subtracting out (reserving) the bridge funding and then computes the potential after loss withdrawal on the remaining portfolio?
In other words, if one used the older approach and funded the bridge in real life with TIPS or a CD ladder, and then suffered a 50% equities decline, the withdrawal risk would be less than VPW indicates because in real life the bridge would not be subject to the decline?
I'm pretty sure VPW does this but I'm looking for confirmation that I correctly understand that math behind the VPW spreadsheet.
But, if we want to compare both approaches, we must do it in a fair manner. Let's do that.
Let's assume that we have a 62 years old retiree with a $1,000,000 portfolio who delays Social Security to age 70 to receive $25,000/year.
Old Approach (separate portfolios)
The retiree puts (8 X $25,000) = $200,000 into bonds (CD/TIPS ladder, or savings account) to bridge the gap in Social Security payments until age 70. We'll call this the bridge bond portfolio.
The retiree puts the remaining ($1,000,000 - $200,000) = $800,000 into a 50/50 stocks/bonds portfolio. We'll call this the VPW portfolio.
Note that the overall $1,000,000 portfolio has, effectively, a 40/60 stocks/bonds allocation at age 62.
The VPW portfolio, though, has a 50/50 stocks/bonds allocation. At age 62, the VPW percentage for a 50/50 stocks/bonds portfolio, in the VPW Table , is 4.6%.
So, at age 62, the retiree withdraws $25,000 from the bridge $200,000 bond portfolio, and ($800,000 X 4.6%) = $36,800 from the $800,000 50/50 VPW portfolio, for a total of $61,800 available for taxes and expenses.
If stocks lost -50% of their value just before the withdrawal, the VPW portfolio would shrink to $600,000. The retiree would withdraw $25,000 from the bridge bond portfolio, and ($600,000 X 4.6%) = $27,600 from the VPW portfolio, for a total of $52,600. In other words, as stocks could easily lose half of their value at any time, at least (($61,800 - $52,600) / 12) = $767/month must be budgeted for optional discretionary spending that could be eliminated without affecting the retiree's comfort.
New Approach (single portfolio)
The retiree puts the entire $1,000,000 into a 40/60 stocks/bonds portfolio (similar to the overall portfolio allocation when using the old approach, to keep the comparison fair) and uses the Retirement sheet of the VPW Accumulation And Retirement Worksheet to determine withdrawals.
At age 62, the spreadsheet suggests to take a $61,120 withdrawal and warns about a $734/month impact for a -50% stocks loss.
The outcomes aren't perfectly identical. The initial withdrawal differs by 1% ($61,800 with the old approach and $61,120) and the monthly impact of a -50% stocks loss differs by $33/month (a drop of -$767 in monthly income with the old approach and -$734 with the new approach).
But, I think it's fair to say that both approaches yield similar-enough outcomes. I don't think that the differences would be noticeable in real life.