Skip to content

Simple and Effective Market Timing with Tactical Asset Allocation: summary and key takeaways

The paper tests a monthly rule that holds either a U.S. equity ETF or a long Treasury ETF, whichever rose more over a recent look-back window. Over the 10 years to 12/31/13 it beat the Ivy 5 moving-average portfolio on total return and trade count, but had a somewhat deeper maximum drawdown.

Glenn, L.A. · 2014 · SSRNSummary updated 20 November 2026
Published2014Working paper
Asset focusStocks+BondsTAA
TypeBothTactical Allocation
Reading time4 minthis summary

Simple and Effective Market Timing with Tactical Asset Allocation at a glance

The question
Can a simple switch between stocks and long bonds, based on relative recent performance, beat the popular Ivy 5 trend-following portfolio?
Dataset & period
SPY and TLT ETFs, daily prices, 10-year period from 1/2/2004 to 12/31/13. Longer test with mutual fund surrogates VFINX and VUSTX over a 25-year period from 1/4/1988 to 12/31/13. Ivy 5 uses SPY, EFA, IEF, IYR and the PCRIX fund.
Method
At each month end the portfolio moves fully into whichever of SPY or TLT had the higher price ratio over the look-back window, and otherwise stays put. The look-back length was chosen to maximize return and limit trades, and the author reports low sensitivity to it over a range of values. The Ivy 5 holds equal parts of five asset funds, each only when its price is above its 10-month simple moving average, with cash earning T-bill rates otherwise. Results are shown for a single start date and for every possible entry date.

Key findings of Simple and Effective Market Timing with Tactical Asset Allocation

  • Over the 10-year sample, total return for paired switching was almost 300%, roughly double the Ivy 5 and about three times SPY buy-and-hold.
  • Paired switching needed 45 trades in the 10-year period, against 70 for the Ivy 5.
  • Maximum drawdown was 18.8% for paired switching and 14.6% for the Ivy 5, so the Ivy 5 had the smaller loss.
  • Using the mutual fund surrogates over the 25-year period, total return was almost 2400%, about double buy-and-hold of the equity fund.
  • Paired switching produced higher end-of-period return than the Ivy 5 for every entry date tested, while the Ivy 5 had smaller drawdowns for most entry dates.

Robustness, as reported in the paper

  • The look-back window was varied; performance was relatively stable between 3 and 5 months, and the number of trades fell as the window lengthened.
  • The rule was re-tested over a 25-year period using mutual fund surrogates, with similar drawdown and trade patterns to the shorter sample.
  • Results were computed for every entry date, not only the first day of the sample.
  • Evaluating every 3 months instead of monthly cut performance noticeably. Evaluating daily with an 84-day look-back lifted CAGR by about 1% but multiplied trades by a factor of 4.

Caveats and limits

  • Slippage is not modeled; the author says it could matter with more frequent trading and leaves it to future work.
  • The ETF sample covers only about 10 years, because the funds are young; the longer test relies on mutual funds that carry short-term trading restrictions.
  • The look-back parameter was chosen on the same data used to report results, so the figures are in-sample (inferred).
  • The Ivy 5 comparison substitutes a mutual fund for the commodity ETF used in its original design.

Items marked (inferred) are our reading of the design, not statements by the authors.

Why it matters

The paper shows how a two-asset relative-momentum switch compares with a diversified moving-average allocation on return, drawdown and turnover. It also shows the trade-off between evaluation frequency and trade count. Practitioners can use it as a simple reference design for tactical allocation.

Questions about this paper

What is paired switching?

It means holding one of two negatively correlated assets, here U.S. equities or long Treasuries, and switching on their relative recent performance.

How does it compare with the Ivy 5?

In the 10-year test it had higher total return and fewer trades, but a larger maximum drawdown than the Ivy 5.

Does the paper include trading costs?

Nominal trading costs are included, but slippage is not modeled.

Where can I read the original paper?

The original is at its publisher: https://doi.org/10.2139/ssrn.2437049. We link to the paper and do not host the PDF.

Follow new quant research

Download MacroFinance market data as CSV or Excel, or pull it by API, and test the ideas in these papers yourself. New summaries are added as we review them.

Create a free account
Summaries are written by MacroFinance from the published papers, in our own words, for information only. They are not investment advice and may omit detail. Papers are not hosted here: each page links to the original at its publisher.