We find that technology's effect on employment varies greatly across manufacturing industries. Some industries exhibit a temporary reduction in employment in response to a permanent increase in TFP, whereas many more industries exhibit an employment increase in response to a permanent TFP shock. This raises serious questions about existing work that finds a labor productivity shock has a strong negative effect on employment. There are tantalizing and interesting differences between TFP and labor productivity. We argue that TFP is a more natural measure of technology because labor productivity reflects shifts in the input mix as well as in technology.