PGP is designed to provide most of the gains of bull markets while significantly reducing losses in protracted bear markets.
- Simple and Rules-based
- Low fee, no commissions
- May be completely in or out of the stock market
- Your own portfolio, not a fund
PGP is a simple, rules-based approach to reduce draw-downs while still allowing for much of the upside of stock market investing. We prefer rules-based approaches because they remove the emotions that can interfere in the heat of an active market. Here we describe the rule and demonstrate how it has performed on the S&P (with dividends) using data back to 1926. It also performs well in other equity markets and I can provide research on that.
Own stocks if the current value of the indicator (such as a market index) is at least as high as the average of the last 12 month-end values. Go to cash (or another relatively safe, short-term investment) when the value of the index is below the average.
The theory behind the rule
In essence, this rule buys into up trends and sells into down trends. The rule is consistent with a body of work on portfolio protection. In Dynamic Strategies for Asset Allocation by Andre Perold and William Sharpe, Financial Analysts Journal, January-February 1995 the authors write “Strategies that ‘sell stocks as they fall. . . ’ . . . tend to do very poorly in flat (but oscillating) markets. But they tend to give good downside protection and to perform well in up markets.” People familiar with the details of call option replication – and options are hedging instruments – will recognize the similarity of buying as stocks increase in value and selling as they fall. There are other methods such as Constant Proportion Portfolio Insurance (CPPI) described by A.F. Perold, Harvard Business School, August 1986. The point of these references is that we should expect this rule to offer protection. The idea for this particular strategy comes from the paper A Quantitative Approach to Tactical Asset Allocation by Mebane Faber, Journal of Wealth Management, Spring 2007.
When won’t it work? Disclaimers, warnings, etc.
PGP stands for “Pretty Good Protection” or “Pretty Good Participation” depending on the direction of the market. The name pays homage to “pretty good privacy”, an encryption/decryption algorithm used to protect computer data. The name was chosen to emphasize that it is not perfect. It is designed to protect against protracted bear markets, not sudden crashes. It is geared to avoid the full brunt of bear markets, not small losses. It is also designed to participate in bull markets. Of course what has appeared to work historically may not work in the future or may work to a different degree. As a final disclosure, this strategy will underperform some of the time, particularly in two scenarios. The first is an oscillating (volatile, sideways) market. In such a market this strategy might be subject to whipsaw. In this environment it would under perform buying and holding stocks, but we would not expect large drawdowns. The second period of weak relative performance would be just after the end of a bear market. The strategy
should be in cash and would not participate in the new bull market until a new buy signal.
How Red Tortoise manages PGP portfolios
We offer U.S. only and global versions of PGP. The U.S. only version is either in a cash or short-term, investment grade investments or an ETF that tracks a broad U.S. market index such as the S&P 500.
The global implementation also allows for investment in both developed international stocks and emerging market stocks. The rule is applied independently to the three asset classes. As an illustration, the U.S. rule would decide whether 60% of the portfolio is in U.S. stocks or is put in the relatively safe investment, the international developed rule would apply to 30% of the portfolio, and the emerging for the remaining 10%. Such as global implementation would lo
Our recommended implementation of PGP includes exposure to global markets via U.S. stocks, developed international stocks, and emerging stocks as represented by exchange-traded funds (ETFs) that track the S&P 500, EAFE and MSCI Emerging markets. The U.S. only implementation rule uses an indicator that tracks the S&P 500 for the rule. The global version adds ETFs that track other indices. The decision to invest in other markets is independent. For example, in one implementation a maximum of 60% (at time of purchase) of the assets will be in the U.S. stock market. That will depend on the rule as applied to the U.S. indicator; 30% will depend on an indicator related to developed international stocks and 10% on emerging market stocks. So depending on the three rules, such an implementation could have (upon purchase) 0%, 10%, 30%, 40%, 60%, 70%, 90%, or 100% equities.
Our implementation can reduce the maximum equity by requiring a minimum allocation (e.g. 20% or 40%) to a safer investment such as bonds to further reduce the risk of your portfolio.
Fees, Costs and Minimums
- Our fee is 0.25% annually which is one-fourth of 1%. This is applied quarterly in arrears.
- Minimum account size is $40,000
- No commissions when we implement through Charles Schwab’s Institutional Investment Portfolio program which requires a minimum of 4% allocated to cash.
- The funds we purchase on your behalf have expenses not included in our fee. We do not receive any portion of those expenses.
Hypothetical Performance Based on U.S. Stocks
Disclaimer: This is a hypothetical (a.k.a. paper) portfolio, not actual performance. Future performance may be worse. These returns do not include investment management fees or commissions. We assume a 0.1% transaction costs for the PGP performance. Past performance is no guarantee of future performance.
Click on the following tabs to see different measures of performance.
Long Term Growth
Returns shown here begin December 1926 and end October 2018 providing a history of 91.9 years.
This table shows returns for PGP both gross and net of a 0.10% trading cost. That cost is reasonable in today’s market although it likely would have been higher historically. The point of this analysis is not to come up with what an investor would have experienced in the past, but to approximate what he or she might expect should we encounter similar markets in the future. Unless otherwise stated, the returns for PGP are net.
While these values might suggest that PGP keeps up with the market over a long period, we believe one should be prepared for long term returns below that of the market. Consider the following chart which shows the growth of $1. While PGP is ahead at the end, the S&P itself was ahead until mid-2001. Further, thanks to the depression, PGP got off to a good start and it wasn’t until 1958 that the S&P caught up.
We also note that the balanced portfolio produced noticeably lower returns than both stocks and PGP.
The tables below show how the portfolios performed in periods the S&P Index was Up (at least zero); when it was Down; and, across All periods. NumPeriods are the number of rolling periods. While protection is important, there are more up periods than down so one should carefully consider the relative importance of performance in up and down markets. The next three rows (SP Index Return, PGP Return, and Balanced Return) are the average of the annual returns. The two Percent Capture rows show the ratio of the portfolio performance to the S&P Index. A negative value indicates the portfolio moved in the opposite direction of the market (i.e., the portfolio was up when the market was down).
|S&P Up||S&P Down||All 1 Yr Periods|
|Number of Periods||819.0||273.0||1092.0|
|S&P Index Return||21.0||-14.2||12.2|
|PGP Percent Capture||78.3||26.8||93.3|
|Bal Percent Capture||69.8||45.6||76.9|
|S&P Up||S&P Down||All 3 Yr Periods|
|Number of Periods||892.0||176.0||1068.0|
|S&P Index Return||14.1||-8.7||10.3|
|PGP Percent Capture||87.5||3.5||99.2|
|Bal Percent Capture||76.9||29.9||83.4|
|S&P Up||S&P Down||All 5 Yr Periods|
|Number of Periods||916.0||128.0||1044.0|
|S&P Index Return||12.4||-5.6||10.2|
|PGP Percent Capture||90.9||-38.8||99.7|
|Bal Percent Capture||79.7||5.9||84.7|
|S&P Up||S&P Down||All 10 Yr Periods|
|Number of Periods||934.0||50.0||984.0|
|S&P Index Return||11.0||-1.7||10.4|
|PGP Percent Capture||95.6||-288.8||98.7|
|Bal Percent Capture||82.9||-126.5||84.6|
These charts show the range of the rolling returns. The rectangle contains one-half of the values. The median value is the horizontal line inside the rectangle The whiskers extend to the lesser of 1.5x the box and the most extreme value; outliers beyond that are shown as points. From these charts, one can see
- PGP has produced a similar median value with more consistency (less extremes) than the S&P Index
- PGP has a higher median return than the Balanced portfolio over 5 and 10 years.
- PGP has had more extreme values over 5 and 10 year periods than the Balanced portfolio but has been less extreme over 1 and 3 years.
Market Down 10+%
The following table shows the 15 periods when the S&P dropped by at least 10%. Here are a few observations:
- PGP does significantly better than the worst drawdowns
- When the drawdown is protracted (Months column is larger), it appears to protect better than short drawdowns
- It didn’t perform well in 1987 drawdown which had a duration of only 3 months. The S&P dropped 2.2% in September and then lost 21.5% in October. Because the September drop was small, the rule did not signal to be out of the market. This is important – PGP should not be expected to protect against sudden, sharp declines. It is more effective against protracted declines.
The following table shows the 16 drawdowns for PGP of at least 10%. In addition, we show how long it took to recover. The lesson from this is that PGP offers some downside protection, but one should still expect losses. We can reduce the risk further by always holding some fixed income. While this should reduce the risk, it will also reduce the return because we expect stocks to outperform bonds.
For comparison, the following tables are the drawdowns of at least 10% for the balanced portfolio.
In practice, investors often evaluate their portfolio by examining its performance relative to a benchmark over the last 1, 3, 5 or 10 years. These four charts illustrate PGP’s return relative to the S&P Index and the balanced portfolio. There are a couple of observations to be made:
- There can be large differences between PGP and the S&P Index
- PGP tends to be much better than the S&P when the S&P Index has been in a long decline
- S&P Index tends to be much better when it rises sharply after a long decline because PGP is still out of the market
- PGP tends to be more stable than the S&P with fewer of the highest or lowest returns on a chart
Note: The latest year is through October 2018.
The first year is a partial year starting December 1926.
How to read the scatterplots. The following scatterplots are powerful images for investigating the relationship between the portfolios (PGP and Balanced) and the S&P Index. The horizontal (x-axis) position of each point (dot) represents the annualized return of the S&P. The vertical shows the portfolio performance. The shading indicates the ending year of the performance with lighter shading used for earlier periods and heavier for more recent. The red line is a reference to the return of the S&P. Points above the line indicate higher performance and points below indicate lower. We’d like to see is something like a hockey stick. When the market is down (points to the left of zero, the gray vertical line), we’d like returns more in line with cash suggesting protection. When the market is up (right side), we’d like to be close to the red line suggesting participation.
10 Year Annualized Returns
5 Year Annualized Returns
3 Year Annualized Returns
1 Year Returns