Another way of staying "time-varying risk-premium", is saying that the risk-premium is predictable. However, that the fact that the risk-premium is predictable does not means that you can make money out of this.
The best two references to understand this are:
- Cochrane (2008) - The dog that did not bark
- Goyal and Welch (2007)
The first tells you what economists mean by equity premium being time-varying or predictable. It basically implies that some variable (or state variable) predicts the equity premium. Cochrane argues that mathematically either dividend growth or returns must be predictable. He shows that the latter is true. Take a look at table (1):
The dividend-price ratio predicts the equity premium. When D/P is high the returns are high.
The second reference (Goyal), shows that equity premium is predictable in-sample but not out-of-sample. So you cannot trade on this predictability - which basically implies that you cannot forecast the ex-post return. Ex-ante we know that equity premium moves with some state variables in the economy (i.e. expected returns are high in recessions) but in practice this cannot be exploited economically.
Edit: Following the comments below here is another good reference by Cochrane. In particular in respect to the comment on whether how can risk-premium be predictable but not profitable to exploit, I make my words, Cochrane words:
Does this mean markets are “inefficient”? Is this an invitation to
“buy low and sell high?” Not necessarily. Time varying risk premia are
possible. Think like an economist, and think about market equilibrium,
not trading opportunities. Prices must adjust to eliminate trading
opportunities. People don’t buy stocks because they’re scared. Why at
some times are they more scared than others?