In my first paper, I construct a novel panel dataset on green and non-green public works in Italian provinces. I use the difference between expected and actual spending to isolate the exogenous variation in green spending. I find a positive and significant local green multiplier of around 0.7. In my second paper, I use geocoding techniques to locate green investments at the neighborhood-level and match it with house prices. I discuss the threats to identification and the limits of my data.