It's not just biofuels that require large energy inputs. In fact one of the central tenants of Peak Oil is that the second half of the worlds oil will require exponentially increasing energy inputs to extract, and still yield declining flow rates. And I don't just mean natural gas, coal, blah. It'll require more raw materials (steel, etc.) in the form of extraction/transport infrastructure like deep sea rigs or roads/pipelines for ever more distant wells.

So it's not just the Biofuel part of the liquids production estimates that are meaningless. It's everything in general, the conventional crude production statistics too. None of these even try to estimate on even a global basis what the the energy overhead for energy production is (the energy the energy sector gobbles up), and thus what the net liquids is (production available for other sectors of the global economy). I hear ERoEI figures of 5:1 for Tar Sands and 10:1 for Conventional crude thrown around but I don't know how accurate these may be. I suppose they give you some idea. But it'd be nice if ERoEI was taken seriously by the EIA and IEA. Although I realise how even the estimates for each nations' consumption are difficult to come to, let alone the consumption figures for each nation's energy sector, let alone complicated things like taking emergy into account in the ERoEI (such as the steel used in a rig for example).

So I think the question "if you took corn ethanol out, what would liquids look like???" is the wrong one. The question should be: "If you deducted the energy+emergy overhead for energy production from gross liquids, what net would you be left with?"

I'd say if you did that we'd have been on a decline in "net liquids' energy" since perhaps 2004. Although I'm no expert. Perhaps someone else could hazard a guess what net liquids' energy graphed from 1850 to 2008 would look like.