Interesting question. Not if energy is used to build things that then build things, which it is. Depending on how long that "chain" is on average, the efficiency difference can compound exponentially.
For example, if you have 3 chains of repeated processing, and each link is made 10% more efficient, the total output is increased by 1.1^3, or about 33.1%
I am not sure what kind of a number you could empirically estimate for the "average chain length" of the physical "tool chain" used to produce modern industrial goods, but it's probably at least 2, which would suggest halving efficiency yields cutting final output four-fold!
And yes, I don't actually have a real idea of the answer because it also depends on many other factors like the degree to which energy is the limiting resource for each individual step, how fast capital amortizes (literally how long a factory part lasts between replacement, essentially), etc. I really have no idea either. But my first back of the envelope initial guess would be more pessimistic than "cut efficiency in half, get half the output"
Source: ....got addicted to factorio last week. Please send help.
It doesn't get mentioned much because only conspiracy theorists use ERoEI as a metric, because it's nearly totally meaningless, and so you need to find an actual scientist with the time and interest to show conspiracy theorists why their own sums are wrong.
2014 link:
https://bountifulenergy.blogspot.com/2014/07/renewables-have...
2018 update:
https://bountifulenergy.blogspot.com/2018/10/reports-of-low-...
General dunking on the whole concept of ERoEI:
https://bountifulenergy.blogspot.com/2016/06/eroei-is-unimpo...
Thanks for these, they are a good way to voice something that I've always felt was a problem with the EROI crowd's thinking.
That said, I think the naive interpretation is wrong. IMO most of the wealth we've been creating since the 19th century is mainly about the task of applying cheap energy to new problems: perfecting power looms, artificial fertilizer, transportation, etc. If the energy becomes twice as expensive, it's not clear that we would only be half as rich. We might in fact be much much poorer than our ancestors were before the cheap energy, because the biosphere (our previous primary source of low-entropy raw material) is in far worse shape than it was in 1800 and needs to support many more people.
One hope is that green energy can actually exceed the EROI of fossil energy. If we can get high-efficiency, long-life (don't have to replace it for 100 years or more) PV, then it should be pretty easy to exceed fossil fuels' EROI. You have to dig more coal for every unit of coal-fired work, whereas you can manufacture a solar panel once and let it do unlimited units of work as long as it exists and the sun is shining.