Growth-Trend Timing (Philosophical Economics’ GTT)

Growth-Trend Timing is a stock market timing rule from the pseudonymous writer Jesse Livermore of the Philosophical Economics blog, published in January 2016. It holds the S&P 500 almost all the time. It steps aside into Treasury bills only when two warnings agree: economic growth has turned negative, and the market is below its 10-month average.

The idea#

Simple trend-following, such as selling when the S&P 500 falls below its 10-month moving average, sidesteps most of a long bear market. Its cost is whipsaws: in an ordinary correction it sells near a low and buys back higher, again and again.

The author noticed that the bear markets trend-following earns its keep in almost all happened during recessions, while most of its whipsaws happened outside them. So he switched the trend rule on only when the economy looked recessionary, measured by growth in real retail sales and industrial production. Outside those times the strategy ignores the trend and stays invested.

How it works#

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

1. Compare real retail sales (FRED series RRSFS) and industrial production (INDPRO) with a year earlier, using the latest published figures.
2. If both are growing, hold 100% S&P 500 (SPY).
3. If either is lower than a year earlier, hold SPY while it is above its 10-month average (a 210-day average here), and Treasury bills (BIL) while it is below.

Each figure counts only from the day it was first published, and as first reported. At a month end the latest figure is usually the previous month's, which is the one-month lag the author built in.

What the backtest shows#

The ETF-era test starts in mid-2007, when the Treasury bill fund begins. Since then the strategy returned about 11.5% a year, against 10.7% for the S&P 500 and 8% for a 60/40 stock and bond mix. Its worst drawdown was about 28%, half the S&P 500's 55%, and its risk-adjusted return beat both.

It lost 5% in 2008, when the S&P 500 lost 37%. In most other years it simply matched the market. Its worst drawdown ran from January to October 2022, when it lost 22% against the S&P 500's 18%, and it took until May 2024 to recover.

Since publication at the start of 2016 it has returned about 13% a year.

Real retail sales begin in 1992, so the test with simulated fund history can only start in 1993. From then it returned about 12% a year against 11% for the S&P 500, with a better risk-adjusted return. It lost 2% in 2001 and gained 1% in 2002, when the S&P 500 lost 12% and 22%.

When it struggles#

• Crashes without a recession signal. A fast sell-off that comes before the growth data turn negative is ridden all the way down.
• Inflation shocks. Real retail sales are adjusted for inflation, so high inflation can make them fall even when spending is holding up. In 2022 that let the trend rule move it in and out of a falling market, and it lost more than the S&P 500.
• Recoveries. Once out, it waits for the price trend to recover, so it can miss the first part of a rebound.

Using it on Tactfolio#

The live strategy above runs these rules on SPY and BIL with FRED data. Each figure is used only from its release, so the backtest never knows a number before investors did. Copy it to try another growth measure or trend length. Compare it with the unemployment version, or with Faber's 10-month moving average, which applies the trend rule all the time.

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

• Jesse Livermore (Philosophical Economics), Growth and Trend: A Simple, Powerful Technique for Timing the Stock Market (2016)
• Jesse Livermore (Philosophical Economics), In Search of the Perfect Recession Indicator (2016)

• The growth test is the headline version from the source: real retail sales (FRED RRSFS) and industrial production (INDPRO) each compared with a year earlier; either one falling turns the trend rule on.
• RRSFS begins in 1992. The source spliced older retail sales series to test back to 1947; this version starts when RRSFS has a year of history, so it has no simulated history before then.
• The source compares the month-end close with its 10-month average; this version uses the daily 210-session average, the closest daily equivalent.
• Each economic figure counts only from the day it was first published, and as first reported: at a month end the latest figure is the previous month's, the one-month lag the source describes.
• SPY and BIL stand in for the source's S&P 500 total return index and 3-month Treasury bills.
• Signals and trades use the close of the last trading day of each month, as in the source.

Common questions#

What is Growth-Trend Timing?#

It is a timing rule for the S&P 500 from the Philosophical Economics blog. It uses a 10-month moving average to leave the market only when economic growth data also look recessionary, and holds stocks otherwise.

Which economic data does Growth-Trend Timing use?#

The version here uses real retail sales and industrial production, each compared with a year earlier; if either is falling, the trend rule applies. The author also tested unemployment, payrolls, housing starts, and earnings, and found the unemployment rate the best single signal.

Why use last month's economic data?#

Economic figures come out a few weeks after the month they describe, so at a month end only the previous month's numbers are known. Using them as first reported, on the day they came out, keeps the backtest honest.

Does Growth-Trend Timing still work?#

Since its publication in 2016 it has returned about 13% a year, with its worst loss in 2022. The live record on Tactfolio shows how it is doing now.

Related strategies

• Growth-Trend Timing with unemployment (GTT-UE)
• Growth-Trend Timing with job growth
• Growth-Trend Timing with housing starts and income
• Lethargic Asset Allocation (Keller’s LAA)