Where tactical-asset-allocation relies on judgement, dynamic allocation is mechanical. A volatility-targeting rule, for example, scales equity exposure so that forecast portfolio volatility stays near a fixed number: if the target is 10% annualised and realised equity volatility is 20%, the rule holds 50% equities; when volatility doubles to 40%, it holds 25%.
The appeal is that exposure falls automatically in turbulent markets, which historically cluster with poor returns. The cost is turnover and whipsaw. A rule that de-risks after a sharp drop and re-risks after a sharp bounce can sell low and buy high several times in one year.
Any dynamic rule should be tested across regimes, not just the last decade, and its trading costs should be included. See backtesting and glide-path for the pre-planned version of the same idea.
Related: tactical-asset-allocation, risk-parity, volatility, glide-path, backtesting