In a linear regression analysis, suppose the independent variable $x$ is measured perfectly, but the dependent variable $y$ is measured with a random error $v$ (so $y^* = y + v$, where $v$ has mean 0 and is independent of $x$ and the regression error term $\epsilon$). If we fit a regression of $y^*$ on $x$, what is the effect on the estimated slope and its standard error?