Evidence for a Lower Value for $H_0$ from Cosmic Chronometers Data?

An intriguing discrepancy emerging in the concordance model of cosmology is the tension between the locally measured value of the Hubble rate, and the 'global' value inferred from the cosmic microwave background (CMB). This could be due to systematic uncertainties when measuring $H_0$ locally, or it could be that we live in a highly unlikely Hubble bubble, or other exotic scenarios. We point out that the global $H_0$ can be found by extrapolating $H(z)$ data points at high-$z$ down to $z=0$. By doing this in a Bayesian non-parametric way we can find a model-independent value for $H_0$. We apply this to 19 measurements based on differential age of passively evolving galaxies as cosmic chronometers. Using Gaussian processes, we find $H_0=64.9 \pm 4.2$ km s$^{-1}$ Mpc$^{-1}$ $(1\sigma)$, in agreement with the CMB value, but reinforcing the tension with the local value. An analysis of possible sources of systematic errors shows that the stellar population synthesis model adopted may change the results significantly, being the main concern for subsequent studies. Forecasts for future data show that distant $H(z)$ measurements can be a robust method to determine $H_0$, where a focus in precision and a careful assessment of systematic errors are required.