We develop a top-down strategic pathway towards green investing and show that eco-investing should not be a puzzle or a sacrifice if investors consider the unknown impact of climate change on different asset classes. We investigate the impact of climate change over a large universe of asset classes including stocks, bonds, alternatives and a list of green assets which have low-carbon emissions, in terms of their time-varying risk-return trade-offs using a factor-based vector auto-regression (VAR) model with climate risk as an additional state variable. Estimation results show that green assets are, in general, resilient against temperature change, while many grey assets are negatively related to the risk of warming. When simulating a future investment opportunity set, we disentangle the climate effects from the return processes and introduce two climate scenarios: an optimistic scenario implied by the VAR estimates; and a pessimistic one based on the DICE model. We find a substantial demand for green assets under the pessimistic scenario, which verifies the hypothesis that investors with a decent awareness of climate change should find it profitable to invest in green.