I doubt there's a generic answer to the question Mr. Schaefer of the Oil & Gas Investment Bulletin poses. It's not just a matter of whether it might be more profitable for these refineries to process heavy oil and export light, as I've suggested in the past, but if refineries that have been modified for a heavy crude slate can still revert to running the lighter crudes coming out of the Bakken and Eagle Ford shales, which seem broadly similar to those for which many of them were originally designed.
To a large extent that would depend on the specific configuration of each refinery, along with the market it serves. Understanding why requires at least a cursory understanding of how refineries work. If you look at even a simple refinery flow schematic, you'll notice numerous interconnections between the various process units. Unfinished products flow from one unit to the next until blended finished products come out the other end, but those aren't the only important flows. What makes today's refineries around 90% efficient is clever recycling of heat and byproducts like hydrogen and fuel gas. In order for the whole facility to operate smoothly, all its units must remain in balance with each other. If you feed the atmospheric distillation column, or "crude unit" at the front end with very light crude oils that, when distilled into their components, overwhelm some downstream units and starve others, that careful integration will break down.
This seems to be the essence of Mr. Shaefer's point: that the Bakken and Eagle Ford crude streams are too light, yielding abundant LPG, gasoline and diesel but insufficient "gas oil" and residue to keep the fluid cat cracker--the gasoline-making heart of a modern refinery--and other downstream units running properly. For some facilities in some locations that could be an important insight and limitation. Certainly the California refinery where I once worked as a process engineer would have choked on 47 gravity Eagle Ford or 43 gravity North Dakota Sweet, having been designed for an average API gravity (a measure of crude oil density) of around 20. However, it's worth recalling that no significant new oil refinery has been built in the US since the 1970s, although many have been upgraded substantially since then. That means that most of today's Gulf Coast refineries have cores that were built at a time when domestic light sweet crude was still abundant.
Refinery engineers and the supply departments that work with them can also be quite resourceful. I recall that when I was a trader for Texaco's US West Coast operations in the 1980s, we routinely purchased intermediate feedstocks from other refiners, to keep our refineries' cat crackers and cokers running during crude unit maintenance or when their current crude slates didn't naturally provide enough. Between shifts in the mix of other crude oils run in conjunction with light oil, and supplementing with feedstock purchases, Gulf Coast refiners might have more flexibility to maximize purchases of shale oil, or "tight oil" than Mr. Shaefer's argument credits.
Ultimately this issue factors into a larger debate concerning how to extract the greatest benefit from the new oil bounty that shale production techniques are providing. Are we better off with constrained pipelines that force some producers to discount their output so it can still reach market by rail or truck, and incidentally provide some refineries with a cheap source of crude? Does the maximum advantage accrue from ensuring that all US crude production is processed in US refineries, even if they increasingly export their products internationally, in light of weak US demand? Or should we suspend outdated crude oil export restrictions and allow both producers and refiners to compete in the same global market that already sets most US refined product prices? That's a worthwhile debate, and I'd be surprised if it were settled based on the physical constraints of the sophisticated refineries of the US Gulf Coast.