GTAA Agg3 by Meb Faber Overview

GTAA Agg3 is the concentrated version of Meb Faber's Global Tactical Asset Allocation. Each month it ranks thirteen global asset classes by recent momentum, holds the top three at a third of the portfolio each, and holds cash in place of any of those three whose price is below its 10-month moving average. It is a rotation strategy with a trend-following brake, built from the most widely read research paper tactical asset allocation has ever produced.

Dual Momentum Systems tracks it because the paper behind it is the reason a great many people have heard of tactical allocation at all. Anything published since owes it something, and a strategy that influential deserves to be measured rather than admired.

The Backstory

The paper nobody wanted to read

In 2006 Faber wrote a draft white paper called A Simple Approach to Market Timing. By his own account, he circulated it among friends and professionals and found that essentially nobody was interested. He retitled it A Quantitative Approach to Tactical Asset Allocation and published it in 2007, in The Journal of Wealth Management.

The core idea was almost aggressively plain. Take a handful of major asset classes, hold each one when it is above its 10-month simple moving average, and hold cash in its place when it is not. Check once a month. That is the entire model. Faber tested it on U.S. equities in-sample and on more than twenty other markets out of sample, then ran it as a five-asset portfolio going back to 1972.

The finding was not that timing made you rich. It was that timing barely changed long-run returns while cutting volatility and drawdown substantially. Equity-like returns with bond-like drawdowns, in his summary.

Then the timing became famous

The paper was published in 2007. The global financial crisis arrived immediately afterward, and a model whose entire purpose was to step aside from sustained declines did exactly that while most portfolios lost half their value.

That timing turned an obscure paper into a phenomenon. It went on to become, by Faber's own account, the most downloaded paper of all time on the Social Science Research Network (SSRN), with roughly 200,000 downloads. For a piece of practitioner research with one rule in it, that is remarkable.

The reasons it spread are worth understanding, because they explain the entire genre that followed. The rule was simple enough to explain in a sentence and simple enough to execute in a brokerage account with a calendar reminder. It was published free, in full, with its data described. It was tested across many markets rather than optimized on one. And it arrived at the exact moment a generation of investors were looking for something other than buy and hold. Faber also updated the paper with new data in 2013 and revisited it again ten years on, which is more follow-up than most published strategies ever receive.

From paper to fund company

Faber did not leave the attention unused. He had co-founded Cambria Investment Management in 2006 with Eric Richardson, and the paper became the foundation of a public profile: the Meb Faber Research blog, The Ivy Portfolio in 2009, later books including Shareholder Yield and Global Value, and eventually one of the more widely listened-to podcasts in finance.

The strategy itself reached the market as an ETF in October 2010, when the Cambria Global Tactical ETF launched on the AdvisorShares platform with Cambria as sub-advisor. It was an ETF of ETFs running a tactical global allocation, and it gathered around $60 million. Cambria and AdvisorShares parted ways in 2014, with the fund passing to another manager.

By then Cambria was issuing its own funds. The Cambria Shareholder Yield ETF arrived in May 2013, its foreign and emerging counterparts soon after, and the lineup kept expanding from there. It is a straight line from a free white paper nobody would read to a fund company: publish the research, let the results do the arguing, build the firm on the audience.

That trajectory is worth noticing on its own. Faber gave the strategy away completely and still built a business, because the value turned out to be in the credibility rather than in the secret.

From five assets to Agg 3

The original model was deliberately unconcentrated: five asset classes, equal weight, each independently in or out. The later expansions added more asset classes and a ranking step, which is where the aggressive variants come from. Instead of holding everything that passes the trend test, rank a broad universe by recent momentum and hold only the strongest few. Agg 3 holds three. Agg 6 holds six.

That is a real change in character. The original was a risk-reduction model. Agg 3 is a concentration model with the original's brake still attached: relative momentum decides what to own, and the moving average decides whether to own it at all.

Why DMS carries it

GTAA Agg3 is the clearest example of a design this site takes a different approach to. It ranks a wide universe and holds the winners, letting breadth come from the size of the universe rather than from the structure of the portfolio. Strategies built here mostly do the opposite, fixing the structure and letting momentum decide within it. Running both on the same data, with the same costs, is the only way to see what that choice actually costs or earns.

Core Strategy Logic

GTAA Agg3 is evaluated once a month. Signals come from month-end data and set the holdings for the following month.

Step 1 - Rank the universe (relative momentum)

For each asset, compute 1, 3, 6 and 12-month total returns and take their plain unweighted average. This is the same 13612 average several Keller and Keuning strategies use. Rank all thirteen assets by that figure.

Step 2 - Hold the top three

The top three each receive a third of the portfolio.

Step 3 - The trend brake (absolute momentum)

Each pick must also have a price above its 10-month simple moving average. If it is below, that slot holds cash instead. The moving average test is the surviving piece of the original paper, and it is what keeps a top-ranked asset in a falling market from being bought simply because everything else is falling faster.

The full portfolio is rebalanced monthly whether or not the selection changed.

The universe as implemented here

Category	Assets
U.S. equities	large-cap value (IWD), large-cap momentum (MTUM), small-cap value (IWN), small-cap (IWM)
International equities	developed markets (VEA), emerging markets (EEM)
Bonds	intermediate Treasuries (IEF), long Treasuries (TLT), corporate bonds (LQD), international Treasuries (BWX)
Real assets	commodities (PDBC), gold (SGOL), real estate (VNQ)

Cash is short Treasury bills (BIL).

Two implementation notes belong in the open.

Ticker substitutions. A few positions use this site's standard fund for an exposure rather than the one in the reference implementation: developed international, commodities and gold all map to the fund used across every strategy here. The exposure is the same; the specific fund is chosen for data quality and consistency.

Assets join the ranking when their history supports it. Several funds in this universe are young. An asset enters a given month's ranking only if both its momentum average and its 10-month moving average can be computed for that month, and is skipped otherwise. The early decades therefore rank fewer than thirteen assets, and the universe fills in as each history begins. This matches the reference implementation, which excludes emerging markets and international Treasuries from its own early years for the same reason. It does mean the strategy's distant history and its recent history are not testing quite the same universe.

Performance Highlights

Over the full published history:

January 1980 through August 2026, net of trading friction:

	CAGR	Max Drawdown	MAR
GTAA Agg3	+12.1%	-20.0%	0.61
Hybrid Asset Allocation	+14.1%	-9.6%	1.46
S&P 500	+12.0%	-51.0%	0.24

Against the index, GTAA Agg3 delivers a similar return with a much smaller worst decline, which is precisely what the original paper promised and a fair vindication of it.

The Hybrid Asset Allocation row is the more interesting comparison, and not a flattering one. HAA came fifteen years later, uses the same momentum average, and is built on the same premise of ranking a broad universe with a defensive filter. Its more careful construction, in particular a defensive test that does not depend on the ranked assets themselves, produces a far better result on both sides of the ledger. That is what fifteen years of public iteration on a published idea looks like, and it is a compliment to the paper that started it as much as a critique of this variant.

The same three, measured from January 2000:

January 2000 through August 2026, net of trading friction:

	CAGR	Max Drawdown	MAR
GTAA Agg3	+10.4%	-15.9%	0.65
Hybrid Asset Allocation	+11.5%	-9.6%	1.19
S&P 500	+8.0%	-51.0%	0.16

GTAA Agg3 holds up well in the more recent window. Its return advantage over the index is wider here than over the full history, and its worst decline is smaller than in the full record, which means its deepest drawdown happened before 2000 rather than in the dot-com or 2008 declines. For a strategy that became famous in 2008, performing respectably in the two decades of scrutiny that followed is the harder test, and it passes.

Two cautions apply as always. A window starting in January 2000 begins shortly before a major bear market, which flatters anything that steps aside during one. And a single figure spanning decades says nothing about the order the returns arrived in, which is most of what an investor actually experiences.

Portfolio Characteristics

• Concentrated by design. Three slots out of thirteen candidates. The portfolio can sit entirely in one category, such as all bonds or all real assets, when that category sweeps the rankings.
• Broad opportunity set. Equities, bonds, commodities, gold and real estate are all eligible, so the strategy can find a trend to follow in environments where an equities-only model has nothing to hold.
• Partial defensiveness. Risk comes off one slot at a time, so the portfolio can be a third, two thirds or fully in cash.
• Trend brake on every pick. Ranking alone never forces a purchase. The 10-month moving average has the last word on each slot.
• Monthly rebalancing. Every month resets the three slots to equal weight, whether or not the picks changed, which means more trading than a hold-until-signal-changes design.
• No leverage. GTAA Agg3 never holds a leveraged fund.
• Best suited to tax-deferred accounts. Monthly rebalancing and frequent rotation generate short-term gains. In a taxable account, plan accordingly.
• Fully mechanical. Every allocation follows from published rules with no discretionary override.
• Scalable. All positions are liquid exchange-traded funds.

Who It's For

GTAA Agg3 is designed for investors who want:

• The best-known name in tactical asset allocation, run as published rather than as reinterpreted.
• Momentum applied across asset classes rather than within equities alone.
• A trend filter on every position, so nothing is held purely because it ranked well.
• Concentration in whatever is working, with the volatility that implies.
• A rules-based process with no market-timing judgment calls.

It is a poor fit for investors who want a stable allocation, who would be unsettled by a portfolio that is occasionally all bonds or all gold, or who trade in a taxable account and are sensitive to short-term gains. Investors drawn to ranking a broad universe but wanting a more developed version of the idea should compare it with Hybrid Asset Allocation. Those who prefer breadth built into the structure rather than into the candidate list should look at Triad.

---

Global Tactical Asset Allocation originates in Meb Faber's research and is presented here as an independent implementation, with the fund substitutions described above. Dual Momentum Systems is not affiliated with or endorsed by Meb Faber or Cambria Investment Management. The original paper is available in full at SSRN.

Past performance, including backtested results, is not indicative of future results. Backtested performance is hypothetical, does not reflect actual trading, and benefits from hindsight in the selection of strategy rules. Portions of the long-term history rely on reconstructed proxy data for funds that did not exist for the full period. Investors should carefully consider their risk tolerance and consult with a financial advisor.

For the latest details, visit www.DualMomentumSystems.com