Ivy Portfolio rotation (Faber’s relative strength model)

The Ivy Portfolio rotation is the momentum version of the five-asset portfolio in Mebane Faber and Eric Richardson's 2009 book The Ivy Portfolio, tested by Faber in his 2010 paper "Relative Strength Strategies for Investing". Each month it ranks US stocks, foreign stocks, bonds, commodities, and real estate by recent returns and holds the top three, each only while it trades above its 10-month average.

The idea#

The book set out to copy, in a simple form, how the Yale and Harvard endowments spread money across very different asset classes. Faber then combined two ideas he had tested separately. Relative momentum says the asset classes that have done best recently tend to keep doing well for a while, so owning the leaders should beat owning all five. A trend filter says an asset below its long-term average is more likely to be in a lasting decline, so each pick is held only while it is above that average.

Together they aim for higher returns than the equal-weight Ivy portfolio with much smaller losses than a fully invested rotation.

How it works#

At the close of the last trading day of each month:

1. Score five funds by the average of their 1-, 3-, 6-, and 12-month returns: US stocks (SPY), developed-market stocks outside the US (EFA), 10-year Treasuries (IEF), commodities (GSG), and US real estate (VNQ).
2. Take the three highest scores, a third of the portfolio each.
3. For each of the three, hold the fund if its price is above its 10-month average; otherwise hold Treasury bills (BIL) in that third.

The 2010 paper averages five lookbacks, adding a 9-month return. This version uses the four-lookback average from Faber's later GTAA Aggressive models, since the builder has no custom multi-period average.

What the backtest shows#

The ETF-era test starts in July 2007. Over that stretch the rotation compounded at about 8% a year, the same as a 60/40 portfolio, with the same risk-adjusted return. The S&P 500 earned more, about 11% a year, but it fell more than half at its worst, while the rotation's worst drawdown was about 21%.

That drawdown came in 2008, from June to October, yet the rotation ended 2008 down only 6% while the S&P 500 lost 37%. It recovered by December 2009. Its best year was 2021, up 34%, and it lost 6% in 2022 while the market lost 18%. In strong US stock years such as 2013, 2019, and 2024 it lagged well behind.

Since the 2010 paper it has returned about 8% a year with a worst drawdown of 15%.

With simulated history the test starts in late 1980, and the picture improves. Over the full period it compounded at about 11.5% a year, close to the S&P 500's return, with less than half the S&P 500's worst drawdown and a much higher risk-adjusted return. It gained in 2000, 2001, and 2002 while US stocks fell each year, and 2008 remained its worst drawdown.

When it struggles#

• Long US bull markets. It holds US stocks as at most one of three picks, so it trails when the S&P 500 leads for years, as in the late 1990s and the 2010s.
• Sharp reversals. Momentum and a monthly trend check both react late when a leading asset class suddenly falls.
• Few assets. With only five choices, a stretch when most of them trend down leaves the portfolio largely in bills.

Using it on Tactfolio#

The live strategy above runs these rules on daily data, with a 210-session average standing in for ten month-ends. Copy it to hold one or two assets instead of three, or compare it with the timing-only GTAA 5, which holds all five, and the buy-and-hold Ivy Portfolio.

Year by year

* Partial year.

The rules as implemented

This is the exact tree Tactfolio runs, rebalanced monthly with signals and trades at the close. Open it to inspect or copy it.

Sources and caveats

• Mebane Faber, Relative Strength Strategies for Investing (2010)
• Meb Faber, A Quantitative Approach to Tactical Asset Allocation Revisited 10 Years Later (Journal of Portfolio Management, 2018)

• Top 3 of the five Ivy asset classes with each pick above its 10-month average, else Treasury bills, as in the 2010 paper's combined system and the 2018 paper's aggressive model.
• The 2010 paper averages the 1-, 3-, 6-, 9-, and 12-month returns; this version uses the 1-, 3-, 6-, and 12-month average of Faber's later GTAA Aggressive models, which has no 9-month term. Momentum is the plain average of the 1-, 3-, 6-, and 12-month total returns, with months counted as 21 trading sessions rather than calendar month-ends.
• SPY, EFA, IEF, GSG, and VNQ stand in for the S&P 500, MSCI EAFE, 10-year Treasuries, the GSCI, and NAREIT, and BIL for Treasury bills.
• Faber compares each month-end close with the average of the last ten month-end closes; this version uses the daily 210-session average, the closest daily equivalent.
• Signals and trades use the close of the last trading day of each month, as in the source.

Common questions#

What is the Ivy Portfolio rotation?#

It is the momentum strategy from Mebane Faber and Eric Richardson's The Ivy Portfolio. It holds the three strongest of five asset classes by recent returns, each only while it is above its 10-month moving average.

What ETFs does the Ivy Portfolio use?#

SPY, EFA, IEF, GSG, and VNQ for US stocks, foreign stocks, 10-year Treasuries, commodities, and real estate, with BIL as cash.

How is the Ivy rotation different from GTAA 5?#

Both use the same five asset classes and the same trend rule. GTAA 5 holds all five at 20% and times each one; the rotation holds only the top three by momentum. The thirteen-asset versions of the same idea are GTAA Aggressive 3 and GTAA Aggressive 6.

Does the Ivy Portfolio rotation still work?#

Since the 2010 paper it has returned about 8% a year. Over the whole ETF-era test it matched a 60/40 portfolio's return with a smaller worst drawdown, but trailed the S&P 500. The live record on Tactfolio shows how it is doing now.

Related strategies

• Adaptive Asset Allocation (AAA)
• Dual Momentum: Antonacci’s Global Equities Momentum (GEM)
• Composite Dual Momentum
• GTAA Aggressive 3 (Faber GTAA AGG3)