Permanent Portfolio Thread

Permanent Portfolio Thread


Harry Brown created The Permanent Portfolio, which is a four-way, 25% split across asset categories of:

Long-term Treasury Bonds
Short-term Treasury Bills

These assets are maintained in a 4-way equal split and rebalanced annually or whenever any asset exceeds 35% of total portfolio or falls below 15%. This rebalancing ensures that you are always selling high and buying low.

The portfolio works because there are a finite number of economic environments, and at least one of the asset classes will thrive during that time period and compensate for the losses in the others.

Deflation: Bonds and cash do well. Gold and stocks do poorly.
Inflation: Cash and Gold does well. Bonds do poorly.
Prosperity: Stocks do well. Bonds and gold do poorly.
Recession: Cash does well. Stocks do poorly.

When the last big stock market drop happened in 2009, the PP gained around 2% because long-term treasury bonds carried the entire portfolio.

There’s another forum (gyroscopic investing) dedicated to the PP, and a great book (which I haven’t read, but I’m told it’s brilliant) by Craig Rowland (CraigR) and JM Lawson (MediumTex). The topic of the PP has come up on Bogleheads many times in the past and CraigR maintained a long thread on it in the past. Bogleheads disapprove of the PP because of the high gold allocation. They make statements such as “only want as much gold as is in their wife’s jewelry.”

I find the PP to be a low-volatity method of investing because despite of the high volatility of underlying assets, the overall volatility is very low. Feel free to ask any questions you have about the PP and I will answer them.

Post your asset allocation

Could you give us a sense of overall PP performance over say the last 10-20 years? How does it compare to a total market index fund/ETF strategy for those with time horizon of 20+ years?


Historically, the PP has performed about 10% per year over the last 30 years. This is similar to a stock index fund. However, the volatility is significantly lower. There have been wild swings in an all stock portfolio over the last 30 years whereas the PP has been a smooth rise upwards.

The last few years have been rough for the PP and great for the stock market but I doubt this will last. The study of economics will tell us what will happen but not when it will happen. The stock market is ridiculously overvalued. Economics tells us that, and tells us that it will pop. But we cannot know when it will happen. No one knows. The PP doesn’t require you to know the future.


Yeah, PP is a reasonable, moderate risk diversified portfolio. It probably does better too if you allocated the gold section to bitcoin in the past few years too :slight_smile:

Which ETFs do you recommend for this? There’s a mutual fund, PRPFX, but I doubt there’s a good reason to pay them when you can just buy 4 etfs yourself and then rebalance them every quarter or year. I guess Motif might be an option as a brokerage where you could put your 4 ETFs into a basket and just buy more of the basket for a single trade and rebalance the basket easily as well.


PRPFX is not the same as the permanent portfolio. PRPFX is an earlier version Harry Brown invented in the 70s/80s that has silver and Swiss Francs in it. At the time, the Swiss government had restrictions on printing more money.

The ETFs I recommend are:

  • TLT for long-term treasuries. However it’s preferable to own them individually. Many large brokerages have free trades of US treasury bonds. No cost to buy and sell. By holding TLT, you are paying an expense to the ETF management. It’s cheaper and safer to hold the bonds directly. The primary reason not to hold them directly would be if you are forced to use an account that only allows ETFs (such as a 401k brokerage account that restricts individual bond ownership).

  • SHY for short-term t-bills. Same caveats as above, and I prefer holding individual t-bills if possible.

  • Gold should be held as gold bullion either directly or in a custodian basis. There’s no guarantee the ETFs have the actual gold and gold is meant as an emergency insurance plan during a global collapse. If you must own an ETF, I would avoid GLD. I like SGOL. However, I really like CEF. CEF is a closed end fund with assets in Canada. They actually have the gold, and when gold is undesirable, like it is today, there’s a discount to NAV so you can get it cheaper than it’s worth. When gold is popular, it trades at a premium to NAV. So if you’re buying when it’s down and selling when it’s high, you might make some money on the NAV discount as well. The downside is that it’s 60/40 gold/silver, not 100% gold.

  • Regarding Stock Index funds, I’d pick one of the big ones that is free to trade with at the brokerage of your choice. Vanguard, Fidelity, Schwab all have their own. I’d stick with either SP500 or US Total Stock Market. I would avoid international market funds or global funds.


Just curious…has PP been outperforming BRK.a when comparing performance for past 10 yrs?


PRPFX does have various forms of gold (about 22%). It was recommended to me several yeas ago. Had a test run for a year or two; not that impressed and ditched it (I know, too short a test period). Just my experience.

Why not use cash with 3-4% annual return (do well in deflation, inflation, and recession as you stated) vs. treasury bonds/bills in one’s retirement portfolio, same impacts? Cash is king as they say.

Note: Just a general comment, not specifically replying to Shandril.


I’d absolutely love to do that! 3% to 4% guaranteed annual return (by the most dominant government superpower) on cash sounds great (in the year 2017). Where do I sign up?


You may not be able to: TIAA Traditional annuity (4%); Fidelity Metlife stable value (3%), numerous 3.5%-4% RCA accounts (Taxable interest).


Really? I find this hard to believe. It looks way too conservative.


This site has some exact numbers. They ran it from the early 70s until 2012 and got 9.6% CAGR:


BRK.A invested 9/29/2007 ending today would be an 8.77% annualized return.
BRK.A invested 1/1/2007 ending 12/31/2016 (10 years ending 2016) - 8.4% annualized return
ishares PP had 10 year CAGR of 6.38% after 2016
vanguard PP had 10 year CAGR of 6.32% after 2016



I think it’s also important to look at standard deviation over that time period. The PP ride is a smooth one. A stock-heavy portfolio can have huge swings. Depending on when you need to access the money, you could find yourself needing it when the market is 30% off highs.


Don’t disagree. Was just trying to answer previous question


That could be a very good argument for a retirement income portfolio. Less so for someone investing 30 yrs from retirement when they have much higher risk-tolerance and ability to ride out short-term stock downturns.

My issue with the PP returns is that they stop at 2011, omitting most of the recent historic bull run to keep in touch with the 100% stock market returns. What I means is that to me, the market trends taking into account the last few years are more important than what happened in the mid-70s, as far as predictor of future performance. I believe investing itself and world conditions have greatly changed in the past 20 years so much so that I’m not sure how relevant 40-yr old data is to current and future performance. So is there data on say the last 20-30 years? Something to include returns before the big tech bubble of the late 90s up to today. That will be biased in favor of stocks the other way I believe considering the current bull run but that’d give a good idea of what the greater differential between PP and 100% stock returns could be.

And also I’d be interested to see the difference between the PP and a 50/50 bond/stock portfolio over the last 20-30 years since that 50/50 mix is close to what is often recommended for retirement income portfolios.


I respectfully disagree. None of the past returns are predictors of future performance. One could argue, as you do, that recent performance is a better predictor. Someone else could argue that performance was earlier is a better predictor because markets are cyclical.

It’s not economically possible for the stock market to grow 30% per year, year after year, forever, unless GDP growth is in the double digits.

That said, you could calculate recent performance of the PP. I haven’t bothered and that website I linked hasn’t been updated. Take SPY (for stocks), GLD (for gold), TLT (for bonds) and SHY (for cash) and backtest over recent years if you’d like. I haven’t bothered because I don’t care what the answer is. I don’t believe past performance is a predictor of future returns.

Also, you mention being able to ride out short-term bear markets due to long investing timeline. That assumes the market stabilizes in the short term. Has Japan ever recovered? I think they were in a 20 year bear stagflation market last time I checked several years ago.


While I agree that markets are cyclical, the length of cyclical cycles is not 40-yrs. More like 4 years. Real estate ones have been 15-20 yr long. Even secular ones are closer to 20-30 years hence my preference for data of the last 20-30 years to anything longer than that.

But my main reasoning is that distant data points lose some relevance due to the changing of world (and therefore market) conditions. In 1971, market participation by the masses was different. And for the record, it was not legal to own gold then (that started in 1974 IIRC). It was an era of people having company pensions, not investing on their own, not up to second data market monitoring and data models. Industries were completely different. Steel, coal, manufacturing were main driving forces behind economies. Financials, services, not as much. It seems very unlikely to me that we’ll return to such a system any time soon. Hence why I think the performance of such fundamentally different economies is less good as a predictor (if any is) than more recent trends.

Obviously, I don’t have a crystal ball but the PP make up is that of 25% risky vehicles, and the rest either conservative to very conservative investments. But for simplicity’s sake call it close to a 25% stock/75% bonds portfolio. How that can perform as well as 100% stocks has me a bit skeptical. The context is that higher levels of risk should bring higher returns otherwise nobody would bother taking on more risk than they need to achieve similar returns.

And the sticky point of that analysis to me, is the end data point that seemed to be arbitrarily picked before a period of great stock returns (which are only 25% of the PP). To me, that window period coincides a bit too conveniently with an extremely good run of gold returns (similar to stocks right now) which the PP has a disproportionate stake in. Those returns on gold have cratered since then but that data is omitted from the returns analysis that stops in 2011. A bit like if you looked at 100% stock portfolio now and declared it the best due to its recent returns pulling it ahead of others.

In fact, if you looked at returns of the PP 1982-present, it lags more conventional stock/bond portfolios by quite a lot. Combination of the shift from very high interest rates to historic lows and great stock returns. Does that make the PP worthless? Totally not. It has historically beat the CPI by 4.5% on average with very few long downturns. So I think it definitely has a place in investing. But my argument is that it’s more useful in something close to capital preservation mode than capital growth. In other words, if you have made your wealth and need to keep it, I think it’s may be a very good portfolio. But if you need to aggressively grow your wealth, its lackluster performance vs. more risky investments is gonna hold you back.


Quick & Dirty from Schwab 5yr and 10yr Annualized performance reports:

Fund 5yr 10yr
SPY +14.2 +7.5
GLD -5.2 +6.6
TLT +2.8 +7.2
SHY +0.5 +1.6

    +3.075% +5.725%

Obviously no rebalancing in this, just the performance of each fund.
5yrs you can see that stocks in this current bull run have been killing it, 10yrs takes the last bear market in and things smooth out considerably. Who knows when the current bull will run out of steam, if you believe it will be sooner rather than later, then it would seem the PP would perform even better near term.

As a very recent retiree, it is food for thought. I have always been close to 100% stock portfolio and know that I need to adjust that.


I think this is where the PP is the most useful. Once you’ve retired, 100% stock is probably too aggressive. You’d have to keep taking out money (selling low) for a while and it may take a long while to recover. For capital preservation during retirement, I think some form of PP may be suitable. There are a few versions, some without the cash component that achieve potentially higher returns without changing the Sharpe ratio. But there’s no getting around the trade off of less growth for less volatility (hence more liquidity).

I’m probably 10-15 yrs from retirement so not pressed for a decision just yet but I can see desiring way more liquidity once I get there.


In isolation, I’d judge returns by the risk-adjusted return (Sharpe ratio) but most people here are investing in retirement accounts and won’t be happy with a high sharpe but low absolute return since they can’t easily lever it up in a qualified account.