Genetic selection for calf feed efficiency could lower rearing costs and environmental impact by breeding animals that meet growth requirements with fewer feed resources. Here we estimated genetic parameters for residual metabolizable energy intake (RMEI) as a calf feed efficiency trait using multiple-trait model estimates for energy sink traits and recursive modeling. Data comprised 4,662 weekly average records for metabolizable energy intake (MEI), average daily gain (ADG), and metabolic body weight (MBW) from 938 Holstein dairy calves collected in one research herd from 2016 to 2024. Multi-trait repeatability models including MEI, ADG, and MBW were fit separately for the pre-weaning (PreW; d 5 to 28 of life) and peri-weaning period (PeriW; d 46 to 60 of life), with age, year-season of birth, scour incidence, trial participation, and dam parity as fixed effects. Heritability estimates for PreW-RMEI and PeriW-RMEI were 0.11 and 0.33. In the PreW period heritability estimates for MEI, ADG, and MBW were 0.32 ± 0.05, 0.20 ± 0.04, and 0.54 ± 0.06 respectively. In the PeriW period heritability estimates for MEI, ADG, and MBW were 0.37 ± 0.06, 0.20 ± 0.04, and 0.51 ± 0.07 respectively. Genetic correlations among MEI, ADG, and MBW were moderate to high in the PreW (0.72 ± 0.06 to 0.86 ± 0.06) and PeriW (0.44 ± 0.12 to 0.72 ± 0.06) periods. PreW-RMEI relative breeding value correlations were favorable with heifer fertility traits, while PeriW-RMEI relative breeding value correlations were moderately favorable with selection indices (Pro$ and LPI) and estimated breeding values for protein and fat yield. These findings highlight that genetic selection could be implemented for PreW-RMEI and PeriW-RMEI as distinct early-life feed efficiency traits in dairy cattle. Future studies should be used to collect more phenotypic data across multiple herds and at different stages of life and estimate proper genetic correlations using multi-trait models.