When b1 is less than 1, the impact of past values (X(t-1)) on the current value (Xt) gradually weakens over time. This means the series eventually stabilizes around a certain mean, making it predictable for forecasting.
If b1 is equal to 1 (unit root), the past value has a full impact on the current value, and this effect continues indefinitely. The series keeps moving in the direction of the previous change, making it difficult to predict future values.