Spatial theories of voting and policy implementation face a key challenge: accurately measuring voters' preferences, politicians' ideologies, policies, and the status quo on one scale. This paper introduces a novel technique to estimate these elements collectively.

Through this new measurement approach applied to state-level data, we reveal significant differences in how various policies respond to voter positions over time.

Policy Responsiveness

Tax Policies: Our findings show tax policy outcomes strongly and immediately* mirror shifts in the median voter's position. This aligns with expectations under the Median Voter Theorem.

* Spending Policies: Conversely, spending policies react much less responsively to the changing median voter position, showing only a slight correlation over time.

Mechanism Analysis

We investigated why these divergent responses occur:

Short-term*: Elected officials' positions do indeed shift significantly in response to the evolving views of their median voter constituency. However, this positional change has minimal direct impact on spending policy outcomes.

Long-term*: Instead, our analysis indicates that sustained changes in overall government ideology (driven by shifts in who holds office over time) gradually influence both tax and spending policies.