I was looking for a graph of the distribution of HbA1c levels, I made one myself in the end from the data in one of the papers linked to below .It shows the distribution of HbA1c in the 14000 non diabetic subjects of the US National Health and Nutrition Examination Survey. I haven't done graphs but ethnicity has an influence and participants with an HbA1c between 4.0% and 4.4% were the youngest; those with an HbA1c of 6.0 to 6.4% were the oldest
There are people on the low side of the distribution but why should anyone aim for the lower rather than the middle area of the distribution ?. These are all non diabetics.
There are pros and cons and lots of unknowns.
Risks certainly go up for various conditions with levels above 5%; it's a slow gradual increase at lower levels but the risk definitely increases. Decreasing levels below 6% don't seem to reduce risks by much though
We know in T1 the risk of hypos and more important to me risk of loss of awareness can increase at lower HbA1cs.
In T2 we know that some trials show increased mortality at lower levels but also know these trials used a lot of medication to reduce the HbA1c.
Some research in non diabetics also show an increase in the risks of some conditions with low HbA1cs . (one found increased risks with below 5% another for some conditions at below 4%.)
It may be some low HbA1cs are a marker of an underlying health problem, anaemia, liver problems and Hep C are mentioned ( low levels may be an effect of the disease)
http://care.diabetesjournals.org/content/early/2012/07/30/dc11-2531.full.pdf
http://circoutcomes.ahajournals.org/content/3/6/661.full#T1
I personally think that given we have diabetes ,and for most of us it isn't going away, then we have to set a level that is achievable, safe and gives us the best quality of life possible.