TSP Overview

As a federal employee, the bulk of my retirement savings is invested in the Thrift Savings Plan (TSP).  The TSP is basically a 401k plan for federal workers and those in the military.  The funds in this plan boast some of the lowest expense fees in the industry.  There are five funds to choose from: three equity index funds, a bond index fund and a special government securities fund.  There are also several lifecycle fund options (L funds) which are composed of the five individual funds.  The 2030 L fund, for example, is geared toward those who will retire around 2030.  L funds automatically adjust over time to become more conservative as the target retirement date approaches.

More information on the individual TSP funds can be found here .

Why not use the L funds?

The L funds are heavily promoted by the TSP.  An investor could simply pick the fund with the closest retirement date and never make another allocation decision ever again .  That sounds pretty good and is probably a good choice for some people.  It is the ultimate passive investment strategy.

Why wouldnt I use this option?  I dont like the L funds because the risk/reward trade-off is poor.  Check out the efficient frontier from the TSP website below.  It is sad how much extra risk is taken in the most aggressive funds for an extra percent of expected return.  The three most aggressive L funds are currently composed of between 70-90 percent equities that are all highly correlated with each other.  This means exposure to huge drawdowns.  Thats only acceptable in my book if the expected returns are great enough.  It can be seen below that expected returns are 7-8-ish percent for the 2020 and later funds.  Sorry, but that just doesnt cut it for the amount of risk taken.  I also hope to earn more than the expected 5.5-ish percent return of the L income fund during my retirement years.

So how can one expect to earn a high rate of return and cut down on risk?  The answer is to use a more active approach.

TSP Strategy Overview

My TSP strategy is based on a technical model that has been hypothetically back-tested with data from the past 50+ years.  Average annual returns between 1958 and 2010 would have been about 10.9 percent with a low 7 percent standard deviation and no years with negative returns. The strategy uses a moving average crossover system similar to ones employed by others.  Allocation changes are made infrequently, maybe once or twice a year on average.  There is nothing magical about it or any of the parameters that I use.  The strategy only invests in two funds the C and G funds.

The goal of the strategy is not to beat the market over the long term.  The goal is to earn a high rate of return with low downside risk. The strategy generally underperforms the stock market during a bull market in stocks.  It destroys the stock market during strong bear markets.   Volatility of annual returns is so low that I would use this strategy at any age.

Strategy Rules

Allocations in the strategy vary between 100 percent G fund, 100 percent C fund, and 60 percent G fund/40 percent C fund.  For those curious, the rules are as follows:

Rule 1 :  If the 7 day simple moving average of the S&P 500 index falls 4 percent or more below its 200 day simple moving average, allocate everything to the G fund.  (Notes:  This acts as a stop-loss that protects from extended drawdowns)

Rule 2 :  After rule #1 has gone into effect, if the 7 day simple moving average of the S&P 500 index rises 2 percent above its 200 day simple moving average, allocate everything to the C fund. (Notes:  This provides maximum exposure to stocks when the model suggests a high success rate)

Rule 3 :  After rule #2 has gone into effect, if the price of the S&P 500 index rises 10 percent  or more from when the signal from rule#2 went into effect, allocate 60 percent to the G fund and 40 percent to the C fund.  (Notes:  This limits exposure to stocks when the model is most unclear yet still allows for significant gains should stocks climb unabated for an extended period)

Performance

Performance of the strategy would have been good on a total return basis and very impressive on a risk-adjusted basis.  About 10.9 percent return over the long run with a 7 percent standard deviation .  Going forward, I hope to achieve at least a 9 percent return over the long run with similarly low volatility.  I want to allow for somewhat lower returns to account for risks Im overlooking and interest rates possibly staying low for an extended period (thus keeping the average G fund rate lower than the past 50 years).