I think it doesn’t matter much, once you compare multiple data points. I’m largely with TeaCup on this one. I also tried to estimate it from the ICTAX figures a while ago (even posted it on the forum), and estimated the difference between VT and VWRL to about 1% every six years or so.
While true, isn’t it a bit pointless?
Realistically, an investor that wants to keep things very simple will choose between VT or VWRL (such as thread title suggest). Whatever little nuances in index composition.
I’ll shout out my opinion:
You’re all getting too hung up on small nuances and overanalysing! 
My take is this:
- It’s safe to assume that a few small caps aren’t going to make a very substantial difference. So VT and VWRL are reasonable subsitutes.
- It’s also safe to assume that a U.S.-domiciled World ETF will be slightly more cost- and tax-efficient (when comparing physically replicating ETFs)
- The difference is not going to be huge. It’s reasonable to assume that it’s more than 0.1% but less than 0.3% per year.
As long as costs are all and everything one care about and one feels compelled out the last fraction of a percentage point in performance, he or she absolutely should prefer VT (though beware of the small tax pitfalls).
As long as you’re comfortable to pay a bit more and don’t lose sleep over a fraction of a percentage point in performance, you can just as well buy VWRL and stop worrying. Also makes the tax declaration (slightly) simpler.