Out-Performing the US Market With the iM-Country Rotation System

  • The iM-Country Rotation System periodically selects one country ETF from six countries, USA, Canada, Japan, Australia, Germany and Sweden, based on the performance of their respective currency ETFs.
  • Backtests from 2009 to 2018 (the bull market period) show that each foreign country ETF under-performed the US stock market over the full backtest period.
  • However, when periodically selecting ETFs using a ranking system based on the performance of the countries’ respective currency ETFs, the model significantly out-performed the US stock market.
  • Over the period 3/9/2009-7/21/2018 the system showed a simulated annualized return of 29.4% versus 18.7% for the SPDR S&P 500 ETF Trust (SPY), with similar maximum draw-downs of about -19%.

Model Philosophy

This model invests periodically in only one country ETF of the six available for trading. The selection is based on the performance of the respective currency ETFs over a one year period.

Country ETF
USA SPDR S&P 500 ETF Trust (SPY)
Canada iShares MSCI Canada ETF (EWC)
Japan WisdomTree Japan Hedged Equity Fund (DXJ)
Australia iShares MSCI Australia ETF (EWA)
Germany iShares MSCI Germany ETF (EWG)
Sweden iShares MSCI Sweden ETF (EWD)

Note: Countries such as China and Switzerland which are monitored by the US Treasury Department for currency manipulation are excluded. Also the United Kingdom is not represented due to the “Brexit” effect on its currency.

Country Currency ETF
USA Invesco DB US Dollar Index Bullish Fund (UUP)
Canada Invesco CurrencyShares Canadian Dollar Trust (FXC)
Japan Invesco CurrencyShares Japanese Yen Trust (FXY)
Australia Invesco CurrencyShares Australian Dollar Trust (FXA)
Germany Invesco CurrencyShares Euro Trust (FXE)
Sweden Invesco CurrencyShares Swedish Krona Trust (FXS)

Ranking System

The performance indicated by the ranking system for the six country ETF universe is shown below. The ranking system ranks the six country ETFs according to the 1-year performance of their respective currency ETFs, lower being better. There are six “buckets” each holding one country ETF. Accordingly, selecting periodically the highest ranked ETF would produce an annualized return of about 29% over the period 2009-2018, which is the same return the simulation produced (Figure-2).

width="640"/(click to enlarge)

Performance of the Country ETFs versus SPY

Figure-1 shows the performance of the five country ETFs representing Japan, Australia, Canada, Sweden and Germany versus the benchmark ETF (SPY). One can see that all five ETFs under-performed the benchmark by a wide margin over the full backtest period. However, there are shorter periods when some of them performed better than the benchmark.

width="640"/(click to enlarge)

Performance of the iM-Country Rotation System

In Figure-2 below the red graph represents the model and the blue graph depicts the performance of the benchmark, the SPDR S&P 500 ETF (SPY). The model is always fully invested, with no market timing.

When ranking country ETFs according to the performance of their respective currency ETFs and periodically selecting the highest ranked ETF, the rotation system significantly out-performed the benchmark (SPY). The simulation shows almost a 30% average annualized return over the preceding 9 years, or a total return 2.5 times greater than for (SPY), with similar maximum drawdowns of about -19%.

width="640"/(click to enlarge)

The iM-Country Rotation System was backtested on the on-line simulation platform Portfolio 123. Trading costs, including slippage, were assumed as 0.05% of the trade amounts using closing prices. Tables 1-4 in the Appendix provide Performance Details, Risk Measurements, Trading Statistics and a listing of the Realized Trades.

Following the Models

A Silver membership is required to follow the weekly performance update of the iM-Country Rotation System. As of writing (7/26/2018) the current holding is EWD since 6/25/18 showing a return of 5.6%.

Disclaimer

The performance data shown represent past performance, which is not a guarantee of future results. Investment returns and portfolio value will fluctuate, so ETFs when sold may be worth more or less than their original cost. Future performance may be lower or higher than the performance data cited.

Appendix

width="640"/(click to enlarge)


width="640"/(click to enlarge)


width="640"/(click to enlarge)


/

Posted in blogs, featured
21 comments on “Out-Performing the US Market With the iM-Country Rotation System
  1. randyfloyd says:

    Really good diversification outside of your existing strategies. Maybe i missed it… is there a safe asset option for when all the markets are tanking?

  2. numejak says:

    What if one only held small cap and cyclical/discretionary ADR’s from the country this model signals, and only during the seasonal months of November through April? Then only defensive/non-cyclical sector ADR’s from the country this model signals from May through October? could this amplify gains of this model? (Essentially we would be applying the Seaonal Rotation approach to this model.)

    • geovrba says:

      We would need those ETFs for each of the six countries used. They are available for the US, but I am not aware of any for the other 5 countries.

  3. smaug says:

    This is a very interesting strategy due to its simplicity. Why are you using the hedged DXJ rather than EWJ, when all the others are unhedged?

    • geovrba says:

      Returns are better with DXJ than for EWJ.

      • TDCARLSON says:

        back of the envelope: looks like returns with EWJ still would have been 25%+

        Tom C

        • geovrba says:

          You are close: model with EWJ return is 25.6%
          Period 03/09/09 – 07/26/18

          Benchmark S&P 500 (SPY)
          Ranking System Currency Basket – with EWJ
          Quick Stats as of 7/26/2018

          Total Return 747.18%
          Benchmark Return 404.96%
          Active Return 342.22%
          Annualized Return 25.58%
          Annual Turnover 433.14%
          Max Drawdown -21.19%
          Benchmark Max Drawdown -18.61%
          Overall Winners (28/43) 65.12%
          Sharpe Ratio 1.26
          Correlation with S&P 500 (SPY) 0.78

  4. danw says:

    I was wondering why you used DXJ for Japan rather than the iShares EWJ (like you did for all of the other currencies)?

    Thanks,
    Dan

  5. prademacher says:

    Why did you choose these 6 countries to be in the model?

    • geovrba says:

      Developed countries only and free floating exchange rate are a requirement. The BRIC countries are not used, nor any of the currency manipulators, such as China and Switzerland.

  6. TDCARLSON says:

    Great performance. I’m sure you don’t have ETF history, but do you have any thoughts on how this might have performed during the big U.S.bear markets of 2001-03 and 2008-09, and on whether timing based on a S&P long/short signal overlaid on this might have helped?

    Tom C

    • geovrba says:

      Drawdowns during US bear markets would be similar to SPY. Best to be out of stocks during those times. We have a special market timer (not on website) which would have kept one out of stocks for only two periods:
      2/5/2001 to 3/31/2003 (784 days)
      12/24/2007 to 3/23/2009 (455 days)

  7. finne00244 says:

    Would be nice to see this model going back further. Looks like the currency etf’s all began trading in 2007. Any chance of running the returns back as far as you can?

    • geovrba says:

      The ranking system ranks the six country ETFs according to the 1-year performance of their respective currency ETFs. The latest inception date for the currency ETFs is 3/1/2007 for UUP, which implies the earliest start date for the model of 3/1/2008.

      Here are the corresponding performance stats:
      Period 03/01/08 – 07/31/18

      Benchmark S&P 500 (SPY)

      Quick Stats as of 7/31/2018
      Total Return 602.35%
      Benchmark Return 160.05%
      Active Return 442.30%
      Annualized Return 20.58%
      Annual Turnover 456.11%
      Max Drawdown -53.80%
      Benchmark Max Drawdown -51.48%
      Overall Winners (32/52) 61.54%
      Sharpe Ratio 0.96
      Correlation with S&P 500 (SPY) 0.84

  8. randyfloyd says:

    Can you please do a backtest in which you overlay the iM Composite Market Timer? In other words, have this model move into IEF when US markets signal an out?

  9. jeremyjmcneil says:

    Hi there – just so I’m clear on this model. The ranking system uses the country currency ETF, sorts that from lowest to highest 1-Yr return, and then buys the country ETF that had the lowest currency ETF return? Do I have that right?

Leave a Reply

With reference to Section 202(a)(11)(D) of the Investment Advisers Act: We are Engineers and not Investment Advisers, read more ...
By the mere act of reading this page and navigating this site you acknowledge, agree to, and abide by the Terms of Use / Disclaimer
Share This