Abstract
The SPT-3G 2018 TT/TE/EE cosmic microwave background (CMB) data set (temperature and polarization) is used to place constraints on an axion-like model of early dark energy (EDE). These data do not favor axion-like EDE and place an upper limit on the maximum fraction of the total energy density $f_{\rm EDE}< 0.172$ (at the 95% confidence level, CL). This is in contrast with ACT DR4 which gives $f_{\rm EDE}=0.150^{+0.050}_{-0.078}$. When combining CMB measurements with measurements of the baryon acoustic oscillations and luminosity distance to Type Ia supernovae, we show that the tension with the S$H_0$ES measurement of the Hubble parameter goes up from 2.6$\sigma$ with Planck to 2.9$\sigma$ with Planck+SPT-3G 2018. The additional inclusion of ACT DR4 data leads to a reduction of the tension to $1.6\sigma$, but the discrepancy between ACT DR4 and Planck+SPT-3G 2018 casts some doubt on the statistical consistency of this joint analysis. The importance of improved measurements of the CMB at both intermediate and small scales (in particular the shape of the damping tail) as well as the interplay between temperature and polarization measurements in constraining EDE are discussed. Upcoming ground-based measurements of the CMB will play a crucial role in determining whether EDE remains a viable model to address the Hubble tension.