Testing the local void hypothesis using baryon acoustic oscillation measurements over the last twenty years

A promising solution to the Hubble tension is a local void that is roughly 20% underdense out to 300 Mpc, as suggested by galaxy number counts in the near-infrared. Gravitationally driven outflows from this KBC void might inflate redshifts enough to solve the Hubble tension, a scenario explored in detail by Haslbauer et al. We obtain predictions for the baryon acoustic oscillation (BAO) observables in their best-fitting void models and in the homogeneous $Planck$ cosmology. We compare these models against our compilation of available BAO measurements from the past twenty years. We find that the quality and quantity of available measurements are best using the isotropically averaged distance $D_{\mathrm{V}}$. Taking its ratio with the expected value in the homogeneous model yields good agreement with unity at high redshift, but a discrepancy appears that systematically grows with decreasing redshift. Assuming independent uncertainties, the 42 considered $D_{\mathrm{V}}$ observations give a total $\chi^2$ of 75.7 for the void-free model, while the void models give only $47.3 - 51.2$ depending on the density profile. This represents a reduction in overall tension from $3.3\sigma$ without a void to $1.1\sigma - 1.4\sigma$ in the void models. The $\chi^2$ differences are smaller when considering measurements of the angular BAO scale or its redshift depth. The void-free model provides the worst fit in almost every case. Overall, our results suggest that recent evidence of BAO observables deviating from expectations in the homogeneous $Planck$ cosmology could indicate a local void, which was motivated by considerations unrelated to BAO data or the Hubble tension.