This paper examines whether government intervention can influence the direction of technological innovation in the auto industry, particularly in response to carbon taxes. Using firm-level patent data from 80 countries, the authors find that higher fuel prices (a proxy for carbon taxes) lead firms to innovate more in clean technologies (e.g., electric, hybrid, hydrogen) and less in dirty technologies (e.g., internal combustion engines). They also find path dependence in innovation direction, where firms' past innovation history influences their future choices. Clean innovation is more likely when firms are exposed to clean innovation spillovers from other firms, reinforcing path dependence. The study uses data from the European Patent Office’s World Patent Statistical database (PATSTAT), focusing on high-value "triadic" patents, which are those patented in three major patent offices. The authors construct firm-specific fuel prices based on the geographic distribution of their patents, allowing for microeconomic analysis of directed technical change. They find that higher fuel prices increase clean innovation and decrease dirty innovation, with path dependence in both directions. Additionally, local knowledge spillovers significantly influence innovation direction, with firms more likely to innovate in clean technologies if their inventors are located in countries with higher clean innovation activity. The paper contributes to the literature on climate change and directed technical change, showing how carbon prices can influence innovation. It also highlights the role of firm heterogeneity and local knowledge spillovers in shaping innovation direction. The authors simulate the carbon tax levels needed for clean technologies to overtake dirty technologies, providing insights into the potential effectiveness of carbon pricing policies in mitigating climate change. The study uses a combination of theoretical modeling and empirical analysis to support these findings, offering a comprehensive understanding of how innovation direction is influenced by fuel prices, past innovation history, and local knowledge spillovers.This paper examines whether government intervention can influence the direction of technological innovation in the auto industry, particularly in response to carbon taxes. Using firm-level patent data from 80 countries, the authors find that higher fuel prices (a proxy for carbon taxes) lead firms to innovate more in clean technologies (e.g., electric, hybrid, hydrogen) and less in dirty technologies (e.g., internal combustion engines). They also find path dependence in innovation direction, where firms' past innovation history influences their future choices. Clean innovation is more likely when firms are exposed to clean innovation spillovers from other firms, reinforcing path dependence. The study uses data from the European Patent Office’s World Patent Statistical database (PATSTAT), focusing on high-value "triadic" patents, which are those patented in three major patent offices. The authors construct firm-specific fuel prices based on the geographic distribution of their patents, allowing for microeconomic analysis of directed technical change. They find that higher fuel prices increase clean innovation and decrease dirty innovation, with path dependence in both directions. Additionally, local knowledge spillovers significantly influence innovation direction, with firms more likely to innovate in clean technologies if their inventors are located in countries with higher clean innovation activity. The paper contributes to the literature on climate change and directed technical change, showing how carbon prices can influence innovation. It also highlights the role of firm heterogeneity and local knowledge spillovers in shaping innovation direction. The authors simulate the carbon tax levels needed for clean technologies to overtake dirty technologies, providing insights into the potential effectiveness of carbon pricing policies in mitigating climate change. The study uses a combination of theoretical modeling and empirical analysis to support these findings, offering a comprehensive understanding of how innovation direction is influenced by fuel prices, past innovation history, and local knowledge spillovers.