Gyroscopic Investing

Hi everyone

I’m currently diving deeper into SWR (Safe Withdrawal Rate) and PWR (Perpetual Withdrawal Rate) using the Permanent Portfolio strategy — the classic 25% mix of stocks, long-term bonds, gold, and cash.

For context:
• SWR (Safe Withdrawal Rate) is the percentage of your portfolio you can withdraw annually (inflation-adjusted) while still making the portfolio last a fixed horizon — typically 30 years.
Historically, for the Permanent Portfolio in the U.S., the SWR has been around 4%–4.5%, and around 3.5%–4% in Europe.
• PWR (Perpetual Withdrawal Rate) is the sustainable annual withdrawal rate that theoretically allows you to never run out of money (capital preserved in real terms).
For the Permanent Portfolio, the PWR is typically lower — around 2%–2.5% in the U.S. and 1.5%–2% in Europe, due to lower expected real returns.

The lower volatility of the PP helps protect against sequence-of-returns risk, but total returns are more modest compared to aggressive portfolios.

I’d love to hear from anyone who has experience actually calculating SWR and PWR with PP.
• Which calculation methods or models do you use?
• How do you handle different time horizons or inflation assumptions?
• And… does anyone have historical charts or datasets comparing the U.S. and European versions of the PP?

Any tips, spreadsheets, backtest results, or references would be greatly appreciated


Cheers and talk soon

I use a spreadsheet to calculate the yearly real gain factors for the portfolio/assets

If one year the PP = +10%, inflation = 4% then 1.10 / 1.04 = 1.0577 real gain factor for that year

Repeated for all years. I store those is the first column (a) for each year

I then create a column for each run/series, perhaps 25 year in each, for sequential start years. I prefer to assume the income is drawn at the start of each year, and being against real (after inflation) values that remains constant, such as 4% (SWR). If column A contains the yearly real gain factors then ...

d4=0.04 (SWR)

b10=1
b11=(b10-$d$4)*$a11
b12=b11-$d$4)*$a12
... etc for 25 years worth
then repeated but shifted one down (copy/pasted column b values)
c11=1
c12=(c11-$d4)*$a12
... etc. for 25 years

At the top, rows a, b I add the offset to the last row for each column, b1=35, c1=36, d1=37 ...etc. up to the last column of data. in row b for each column I then use =indirect(address(b1,column()) ... that loads that cell with the last value for each column (final inflation adjusted portfolio value remaining after 25 years of 4% SWR in this case).

That's using Libre Calc, but I suspect it would also be the same/similar for Excel, and in practice my actual columns/rows start more into the worksheet, such as AJ11 in the above image of my worksheet and the SWR value in cell $AH$11 (the dollar prefix as part of the cell value locks the value to that column/row, so when copy/pasted it always still points to cell AH11).

Personally from a UK perspective I see 3% as being a PWR, still had residual portfolio value after 50 years of 3% SWR. At least since 1939. Using 25% USD in US stock (S&P500 total returns data from Simba's backtest spreadsheet), 25% silver pre 1975, gold since 1975, and 50% UK cash rates







50/50 Pounds/foreign, where foreign = 50/50 USD in US stocks and gold (non fiat commodity currency). A multi-way neutral currency/asset hedge (Pounds hedge foreign, foreign hedges Pounds, stocks hedge gold/gold hedges stocks ...etc.).

Yields a lower average reward than other more aggressive asset allocations, however much of that higher 'average' is driven by relatively few great case outcomes, that typically had start dates following recent large/fast declines. In a large majority of cases the 30/whatever year SWR outcomes (residual real portfolio value remaining) of individual (start year) cases weren't that dissimilar whether you were conservative (PP) or aggressive (all stock). A bit like having a room of 10 individuals each of 5 foot height having a 7 foot giant enter the room, the average height is uplifted, however most are below average (the average individual case doesn't achieve the broader average outcome).

The PP's tendency for low drawdowns means that if you start with a PP but reserve the option to move into aggressive at a later stage after large drawdowns have occurred, then your final 30 year outcome can compare or excel that of all-stock from the start, but in some cases will be lower. i.e. PP at the start, transitioned to all stock later after declines and the all stock from the start might have pulled ahead of the PP initially, but later declined back down to compare to the PP, such that a PP transitioned to all stock at that time will end up with the same overall total returns as having been all-stock from the start. Often after a number of years, say 5 to 10, actual portfolio rewards will have supported a higher SWR than the 3% or 4% being drawn, will have grown in real terms such that moving to all stock is less of a risk, as the ongoing SWR value might be just 2% of the portfolio value at that time.

Potential to reduce earlier years bad sequence of returns risk (reduce downside risk), potential to compare or maybe even exceed all stock rewards after 25, 30, whatever years (maximise upside potential rewards).

How do you handle different time horizons or inflation assumptions?

In the UK case that's easily covered by back-testing against long enough time, that include historic cases such as world wars, high inflation/IMF bailout, 2000 dot com bubble burst, 2008 financial crisis, 2020 Covid ...etc. If the asset allocation got you through those periods relatively OK then it might hold up in the future when crises occur. The worst case is perhaps the total loss/decline of the domestic currency - hyperinflation, in which case foreign (USD/US stocks) in domestic currency terms will typically soar - even if they remain relatively little different in price in foreign currency (USD terms), ditto gold. The 50% foreign (USD/gold) hedge the domestic currency. Similarly if the domestic currency relatively rises the 50% in that hedges the foreign currency holdings. Neither fully wrong or right, worse than having been all in the one that did well/better than being concentrated into one alone that falters/fails.

For the Euro you're restricted to it infancy. You'd have to go back to when each of the member states used their own/different currencies, and domestic inflation rates. When however the PP worked well in Japan since 1972, the UK and US since WW2, and others such as Canada, Australia etc. then you might assume the similar may also hold for others, supported a 3% 30 year SWR in all such cases, but where 4% is closer to the wire (reducing SWR even by a little can make a massive difference in outcome). But that is with a common denominator of USD/US stock being held for the 25% stock element, that along with gold creates a 50/50 domestic (50% in bonds)/foreign (USD stocks/gold) balance/barbell.

Let's think on 3% as SWR for the FIRE