A school district assigns students to reading interventions based on a pre-test cutoff: below 50 → Program A (intensive), 50+ → Program B (standard). The outcome model (post-test vs. pre-test) fits as:

• Below 50: $\text{Post} = 0.85 \times \text{Pre} + 24$
• At/above 50: $\text{Post} = 0.85 \times \text{Pre} + 19$

In a regression discontinuity design, the estimated local treatment effect (Program A vs. B) at the cutoff is: