A hallmark of modern humans is that our newborns are neurologically immature compared to other primates. It is disputed whether this so-called secondary altriciality evolved due to remodelling of the pelvis associated with bipedal locomotion, as suggested by the obstetrical dilemma hypothesis, or from maternal energetic limitations during pregnancy. Specifically, the ‘Energetics of Gestation and Growth’ (EGG) hypothesis posits that birth is initiated when foetal energy requirements exceed the maximum sustained maternal metabolic rate during pregnancy at around 2.1 × basal metabolic rate (BMR) of the non-pregnant, non-lactating condition (NPNL). However, the metabolic threshold argued under the EGG framework is derived from one study with a small sample size of only 12 women from the UK. Accordingly, we performed a meta-analysis of all published studies on metabolic scopes during pregnancy to better account for variability. After excluding 3 studies with methodological issues, a total of 12 studies with 303 women from 5 high- and 3 low-income countries were analysed. On average, pregnancy was found to be less metabolically challenging than previously suggested. The studies revealed substantial variation in metabolic scope during pregnancy, which was not reflected by variation in birth timing. Further, in a third of the studies, the metabolic rates exceeded 2.1 × BMRNPNL. Our simulation of foetal energy requirements demonstrated that this metabolic threshold of 2.1 × BMRNPNL cannot realistically be crossed by the foetus around the time of birth. These findings imply that metabolic constraints are not the main limiting factor dictating gestation length.