I know this is off topic, but now I'm curious. Do people who are incapable of taking the test still impact the scores? Does a 100 IQ indicate the median score of the set of "successful" test takers, or of the set of "functional humans", or of the entire population of all humans?
I believe you're saying that IQ 48 is approximately the minimum required level of functionality required to successfully take the test, but there is obviously a set of people who can't achieve that. And while 48 may be the lowest point on the curve that can be measured, the continuation of the curve is still implied below that point. People below 48 will still fall along some spectrum of abilities, but they're not measurable using the current test. So there may very well be someone with an "equivalent IQ" of 14; it's just the current IQ test lacks the resolution needed to identify that person.
And I'm not saying we should expend any effort to alter the test to measure lower IQs. I doubt that would add any value to society, nor would it be likely to benefit the people who can't take the test today. Such people are already identifiable as requiring a certain level of care, and most of the disabilities at that point are so profound you probably couldn't even use the scores to predict the costs of caring for them.