Thank you all for the thoughtful replies.
I'd like to pose another question. I wonder if anyone has compared the magnitude of the tax savings to the turnover and rebalancing costs. Unlike David, I'm only slightly overweighting VSS and VWO, so the bulk of my international will be either Total International or VEA+VSS+VWO. I presume that the slicing approach will suffer more turnover costs over 25 years, and David already mentioned the rebalancing costs. I'm not savvy enough to know if one effect dwarfs the other or if it's a wash. (I'm willing to assume the Total International ER may come down over time)
Per the bogleheads wiki, for the top tax bracket, Total International tax costs (2004-2011 averages) are 0.36% / 0.85% depending on whether 2012 or 2013 cap gains taxes apply. VEA tax costs (2004-2011 averages) are 0.17% / 0.85%.
Per Swensen's 2005 book, For a hypothetical 100% annual portfolio turnover of the largest 500 securities, you would lose 0.5% of the principle in market impact. If we guestimate that the sliced portfolio turnover is at most 20% per year, you might guestimate 0.1% market impact (although I doubt it's a linear relationship) which wouldn't seem to be enough to outweigh the tax benefit with 2012 tax rates. But I have no idea how to model this. Help?