This paper introduces a novel framework for regression with deferral, where the learner can choose to defer predictions to multiple experts. The authors address the unique challenges posed by the infinite and continuous nature of the label space in regression compared to classification. They present a comprehensive analysis for both single-stage and two-stage scenarios, where the learner either simultaneously learns the predictor and deferral functions or uses a pre-trained predictor with a learned deferral function. New surrogate loss functions are introduced for both scenarios, and their consistency guarantees are proven to be stronger than Bayes consistency, being non-asymptotic and hypothesis set-specific. The framework is versatile, supporting multiple experts, any bounded regression losses, instance-dependent and label-dependent costs, and both single-stage and two-stage methods. The single-stage formulation includes the recent *regression with abstention* framework as a special case. Extensive experiments demonstrate the effectiveness of the proposed algorithms.This paper introduces a novel framework for regression with deferral, where the learner can choose to defer predictions to multiple experts. The authors address the unique challenges posed by the infinite and continuous nature of the label space in regression compared to classification. They present a comprehensive analysis for both single-stage and two-stage scenarios, where the learner either simultaneously learns the predictor and deferral functions or uses a pre-trained predictor with a learned deferral function. New surrogate loss functions are introduced for both scenarios, and their consistency guarantees are proven to be stronger than Bayes consistency, being non-asymptotic and hypothesis set-specific. The framework is versatile, supporting multiple experts, any bounded regression losses, instance-dependent and label-dependent costs, and both single-stage and two-stage methods. The single-stage formulation includes the recent *regression with abstention* framework as a special case. Extensive experiments demonstrate the effectiveness of the proposed algorithms.