Investors who live in California pay high income taxes, so CA tax-free bond funds seem worth considering as part of a diversified portfolio. Historically these have demonstrated low volatility of total returns compared to many other asset classes. However, they have been susceptible to occasional drawdowns near ‑10 % when investor sentiment changes. These drawdowns have not occurred overnight; they have taken several months to reach bottom. At present, yields are low, so that a ‑10 % drawdown would wipe out a few years of dividends. I wanted to know if a simple timing strategy could switch from CA tax-free bond funds to cash and avoid the worst of the drawdowns.
Selecting funds to study
I used a Fund Screener at www.schwab.com to select a group of funds to study, as follows:
- Morningstar category = Muni California Intermediate (46 matches). Screening for intermediate- rather than long-duration funds seemed reasonable because of widespread belief that interest rates will rise.
- Open to new investors (22 matches).
- Fees/loads = No-load funds at Schwab (still 22 matches). This criterion seemed important to add because a timing strategy could incur recurring load charges.
- Morningstar overall rating = 4 or 5 stars (9 matches). This reduced the list to fewer than 10 funds, a number I could easily study.
Here are the screen results:
What about minimum initial investments? SICYX requires $1,000,000; EICAX requires $100,000, and the others require $3000 or less. I took SICYX off the list, leaving eight funds.
Brokerage and fund fees
Schwab has a $79.95 transaction fee for EICAX or VCAIX, and it has a short-term redemption fee of $49.95 for any of the other six funds held 90 days or less. Three of the funds have additional redemption fees for large investments: DCAAX has a 1 % fee for investments of $1,000,000 or more held 12 months or less, MECMX has a 1 % fee for investments of $250,000 or more held 18 months or less, and NACMX has a 0.75 % fee for investments of $500,000 or more held 12 months or less.
I used a market timing model tester at www.portfoliovisualizer.com to test a strategy of switching to cash when the excess return of a fund over cash was recently negative. The strategy traded monthly at the end-of-month price. It calculated excess return as the average of the 10-day and 3-month excess total returns.
Backtests looked at three periods: the financial crisis and recovery through 2008 and 2009, the next five years (2010 through 2014), and recent performance (2015 through June 2017).
Sample equity curves
Here are sample equity curves for one of the funds over the three backtest periods. The “timing portfolio” represents the timing strategy described above applied to just one fund, and the “equal weight portfolio” represents buying and holding that fund. The results do not take into account any brokerage and fund fees. They show nominal returns, not inflation-corrected returns.
The equity curves showed that the strategy worked well over the three periods, at least for TECFX.
Backtest performance across selected funds
It’s not easy to generalize from 23 equity curves for the eight funds, so here is another approach. The “risk-return” charts below show compound annual growth rate versus annualized standard deviation, for each fund over each of the three backtest periods. Circles represent the timing strategy, and crosses represent buying and holding the fund. With this type of chart, it’s usually desirable to be in the upper-left quadrant (lower volatility and higher return) compared to the lower-right quadrant.
The chart above shows that during the financial crisis and recovery, the timing strategy was helpful for ALCAX, DCAAX, TECFX, and MECMX, but it was harmful for PBCAX, NACMX, and VCAIX. EICAX did not trade over this full period, so it is not shown.
The chart above shows that in the next five-year period, the timing strategy was helpful for all eight funds, for reducing volatility, increasing total return, or both.
This chart shows that in the most recent backtest period, the funds were closely grouped, and the timing strategy was helpful for all eight funds.
The tables below provide details. They give the compound annual growth rate (CAGR), annualized standard deviation (ST DEV), maximum drawdown (MAX DD), and something similar to the Calmar ratio (CAGR / MAX DD); over each of the three backtest periods.
There is always a risk of overfitting with backtests, although this is a fairly simple strategy with only two parameters. Investors might want to try varying the 3-month and 10-day lookback periods to check for robustness. Investors who live in high-tax states other than California could also try the method above and see how it would have worked in their state.
Different investors have different risk-return preferences. To my eye, it looks like EICAX, MECMX, and TECFX are good choices for this timing strategy. Although results aren’t available for EICAX during the 2008 financial crisis, it had outstanding returns in 2009. It’s also important to keep in mind the 1 % redemption fee for MECMX investments of $250,000 or more held 18 months or less.
Disclosure: I/we have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.
Additional disclosure: I am not an investment professional, and my article does not contain investment advice. Please verify my calculations and make your own conclusions about whether any strategy or security mentioned in my article is suitable for you.