This paper characterizes long-run and short-run optimal fiscal policy in the labor selection framework. Quantitatively, the volatility of the labor income tax rate is orders of magnitude larger than the "tax-smoothing" results based on Walrasian labor markets, but is a few times smaller than the results based on search and matching labor markets. To understand the results, we develop a welfare-relevant analytic concept of "tightness" for the selection model. This concept of tightness is the source of the decentralized economy's inefficient wage premia between the average newly-hired worker and the marginal newly-hired worker. Compared to the well-known concept of "labor-market tightness" in the search and matching literature, this new concept of tightness plays a highly similar role, and, like in the matching model, is crucial for understanding efficiency and optimal policy.