There's a difference between liquidating a fund position because one has lost faith in the fund and adjusting the holding because of performance. (Part of the original question included the example: "sell 2
5% of holding for each 20% gain in a year".)
Performance based adjustments can be done mechanically, based on one's target allocations.
I generally concur with observant1's approach, though I'm more inclined to let a "loser" ride longer, say three years. How much history I use depends on how the fund is managed.
If a fund has a distinctive style, I'll tend to give it more slack. One reason is that it would be difficult to replace. Another more important reason is that because of its style, it may be more likely to do better, or worse, over extended (multi-year) periods.
Here's a good exercise, given that "everyone" thinks M* should have downgraded TPINX before now. When would you have sold it, and why?
The fund had great years through 2010, so let's look at the past decade. Here's a M* page with that data. Pay attention to the benchmark index (world gov bond index) rather than the category returns since the fund was not in that category until recently.
http://performance.morningstar.com/fund/performance-return.action?t=TPINXIn relative terms it was only in 2017 that performance began to fall apart. While it beat its index by 2½% in 2018, it underperformed substantially in 2017 (-
5%+), 2019 (-
5%+), and hugely in 2020 (-14½%).
After its great 2012, in 2013 and 2014 the fund returned very little (2%, 1½%). Would you have sold even though on a relative basis it did great (2013) and average (2014)?
Would you have sold at the end of 201
5 after those two low return years followed by 201
5 when the fund landed squarely in the middle of the pack and fell just short of its benchmark?
Surely you would not have sold after 2016, which was a fine year (6%+ vs 1.6% for its benchmark).
Would you have sold after 2017 which was the first really clear bad year on a relative basis? Or would you have waited to see what would happen?
If you did wait, would you have felt comforted by the 2018 performance when the fund again beat most of its peers and beat its benchmark by over 2%? Or would you have looked at the absolute performance of 1.27% and said to yourself: this is even worse than 2017 where it returned just 2.3
5%. I don't care about relative performance, I'm out?
After 2019's relative disaster, would you have called it quits, perhaps because two of the previous three years (2017, 2019) were very bad (each
5%+ under the benchmark)?
Or would you have waited for two successive bad years relative to its benchmark? It took until 2019-2020 for that to happen.
