Interest Rate Market Timing Model Still on Sell
Posted: August 29th, 2009 | Author: admin | Filed under: Uncategorized | Tags: 401k, Market Timing, Market Timing System, Moving Average, Rate Model, Spy, T Bills, Timing Model | No Comments »This week I am going to add an interest rate model to our market timing arsenal. This particular market timing model looks at the yield differential between ten-year notes and 13-week T-bills. I am not an economist but it appears the market performs better when the differential is declining. Most likely this would be caused when the t-bill rate is increasing relative to yield on ten-year notes. This might happen if the economy was running on all cylinders, and the Fed was raising interest rates in order to slow down the economy. The opposite might be occurring if the differential were rising. The economy could be in the doldrums and the Fed is easing interest rates in an effort to jump-start the economy.
I looked at differential at the end of each month. It is more the trend in the differential that we are interested in, so I do not feel it is necessary to look at the differential on a daily basis. I placed both a 9-period and a 26-period moving average on a chart of the interest rate differential. I would sell the SPY and go into cash if 9-period moving average moved above the 26-period moving average. I would go long the SPY when the 9-period moving average moved below the 26-period moving average. This market timing system only gave three round turn signals in the last ten years. $10000 invested in SPY on 7/31/2000 (the date of the first buy signal) and held until 7/31/2009 would be worth $6349.01. The interest rate model taking only long signal (because I can’t go short in my 401k) would have grown from $10000 on 7/31/2000 to $13,024.27 on 10/31/2007. This is the date of the last sell signal and does not include dividends while in the market or interest earned while out of the market. This means a buy and hold investor, even after the nice up move from the March lows, would still be looking at a drawdown on his or her account of at least 51%.
No market timing system is perfect and this one is not either. Of the three buy signals given since 2000 only two were winners. This system is still on the sideline so it has not caught any of the move from the March lows. This is why I prefer to track multiple market timing models at the same time and diversify my investments among the various timing models.