ETF Allocation by Relative Strength - Faber Model

Here’s the spreadsheet: Simplified Main Strategy - Google Sheets

What is this? This is a monthly reallocation strategy among 14 ETF’s based on each ETF’s relative strength. It also uses FRED economic data to confirm sell signals in recessionary environments.

Where did this come from? Meb Faber’s tactical allocation strategy paper, updated for relative strength. It’s a quick and easy read (see links on spreadsheet main page). The confirmation sell signals in recessionary environments come from a paper on philosophicaleconomics.

Why is the spreadsheet useful? It auto-updates daily, computing each ETF’s relative strength. It also automatically updates dividend and economic data when FRED posts data and when dividends are posted. This should save you time and simplify your investing. It does this using google finance functions and data import functions.

How do I use it? Instructions are on the spreadsheet main page. Every month ETF allocation shifts based on relative strength. Best used in a tax-advantaged account.

How has this performed? See Chart 2 here for a look at historical performance (using a more conservative top 8 instead of top 6 allocation by relative strength). Also note that the Faber model alone does not incorporate FRED data to confirm a sell signal: Extrategic Dashboard - Faber Tactical Asset Allocation w/ Relative Strength (Updated).

1 Like

This is intense… did you make it!?

Valid analytical rigor in its assessments, although I wonder about its dependency on where stocks have trended vs where they might end up.

Example - current ranking has commodities #1 and gov/commercial bonds afther that. Commodities have rallied recently, but if recession happens that wont last. Bonds have been hit hard, and have potentially a lot more to fall as interest rates continue to rise (inverse relation).

The 200 day decline mark it uses may be a smart way to limit losses, although I wonder how that limits potential upside.

All that said, I like how it uses hard FRED data on fundamentals, which means it should be a decent guide to stock performance (bonds removed). I’ll take a deeper dive into this over the next few days.

Yeah I made it as a practical application of the Faper paper a few years ago, then added the FRED data after slogging through the discussion in the online philosophicaleconomics papers (terrible formatting but good analysis).

The most relevant section of the philosophicaleconomics article (
http://www.philosophicaleconomics.com/2016/01/gtt/) is below the sub-heading: " An Improved Trend-Following Strategy: Growth-Trend Timing"

The goal of trend-following strategies like this is to limit drawdowns but otherwise maximize time in the market. The 200-day trendline that Faber ultimately selected does a good job, but the philosophicaleconomics papers add economic data as confirmations to sell signals in similar trend-following strategies. By combining the two concepts, we have something that has not been back-tested (fair warning). You can always ignore the economic data and sell based on relative strength and 200-day trendline alone if you want to go pure, back-tested Faber.

My goal with the spreadsheet is to remove my brain from the equation as much as possible. The more I think about trading, the more time I spend out of the market, and the more taxes I pay.

Skimmed through the article - all solid points, and you can’t argue that momentum investing (of which this concept is an advanced/fully-fleshed out version of) has solid mathematically proven advantages.

Seems like it would absolutely help prevent some long decline losses (i.e. a company failing over years) with its almost stop loss characteristics, but still would be susceptible to large move macro conditions. The inclusion of FRED data might help mitigate it somewhat, but only for moves which are well telegraphed (and therefore largely already built into stock prices).

I like the concept - is there a ETF which executes on this concept? Would much prefer a one-stop trade compared to constant rebalancing/tracking. There are plenty of momentum ETFs out there, I imagine one might come close.

One thing that worries me is that if investing in stocks depends purely on math/data/equations… then we (the little guy) will always be at a disadvantage compared to Wall Street with full time PhD staffs with virtually unlimited ability to source data and compute it. Its why I believe one of the few advantages individual investors may ever hope to aspire to is the intangibles of a company - its leadership, its product, the changing macro conditions. Think Tesla, Netflix, etc. long before they were ever the titans they are today.

Do you use any dividend screener ETFs? Slightly different concept, but I’ve become very interested in ones with SCHD/SCHY characteristics (the second in particular is priced well with overseas weakness right now).

1 Like

Yes and no: Faber’s Cambria Funds has a momentum-weighted trend-following etf, GMOM, but I don’t know how closely it tracks the strategy presented in his paper.

And no, I don’t have a dividend-focused ETF strategy but should look into it.