Are IR renormalons a good probe for the strong interaction domain?

We study the origin of non-analyticity in \alpha_s of a short-distance QCD observable to demonstrate that the infrared renormalons, the same-sign factorial growth of the perturbative expansion, is a universal phenomenon that originates entirely from the small coupling domain. In particular, both the position and the nature of the singularity of the Borel transform of the perturbative series prove to be independent of whether the running coupling \alpha(k^2) becomes singular at some finite scale ("Landau pole"), or stays finite down to k^2=0. We argue that getting hold of the infrared renormalons per se can help next to nothing in quantifying non-perturbative effects.