I can't answer for AD and I'm interested to see what he has to say - but in the meantime this is my take on the situation.
A 5 point harness in a RF seat is a Good Thing as, as Liopleurodon says, it keeps the child where they're supposed to be so that they are as safe as possible in the event of a heavy impact.
A 5 point harness in a FF seat is a Bad Thing for a young child, as:
- The head of a young child is proportionally very heavy
- The spine in a young child's neck is more flexible
- Restraining the shoulders places means if the head is thrown forward in a heavy frontal impact, all of that energy must be contained by the neck. In a young child, the flexibility awarded by young joints means that the spinal cord can wind up taking the strain...
So for a young child I can see an argument that says a 3-point seatbelt is safer in a forward facing seat than a 5 point harness would be - the extra motion it gives allows the body to dissipate some of the energy, giving the neck some protection. Its the same logic that lead to the the Kiddy Infinity Pro car seat. I do think this is where great care needs to be taken, though, to make sure that the child is of an appropriate height and weight to be safe in a 3 point seatbelt, and that the child understands why wriggling around too much could be dangerous (easier said than done, I know). They will be thrown around more in the car with a 3 point harness - its a trade off between allowing the energy of a collision to be spread over much of their body (throwing it around) and allowing a weak point in their body to absorb most of the energy. Maybe this is the point where statistics need to go beyond straight mortality - do more Swedish children come away from accidents alive but with injuries to other parts of their bodies?
(I also think there are issues around whether booster cushions are safe at all - especially when not IsoFIX - and whether IsoFIX makes high back boosters safer as well (ignoring correct fitting) but I haven't had to think about those enough to have a thought out opinion yet!)
A racing car driver is in a bit of a different circumstance - the speeds and forces involved are much higher (meaning they're more likely to be thrown out of the car altogether if they're not held in very securely), and they have a mature skeletal structure in the neck. There is also a huge amount of design that goes into the seat and the cockpit of a racing car to make it as safe as possible, not all of which can be transferred into family cars and/or children's car seats. What is safest for a racing car driver is not necessarily safest for an ordinary motorist - especially not a young one.
Why are seats sold in the US that harness to higher weights? I don't know. Why does the UK government still state on most of their literature that "Group 1 seats are forward facing"? I don't know. A lot of crash tests (including the European standard which the UK uses) do not test for force on the neck - in which case restraining the child's body so it can't be flung around becomes perceived as beneficial (especially when companies like Which? then use the same basis for a test, perform it at higher speeds and publish comparative results). 5-point harnesses are perceived as safer by parents, so they ask for them and manufacturers see a market and make the seats. Are they actually safer? According to some tests, probably they are. According to standards that consider stress on the neck? I don't think they can be.
So I don't know if any of that makes sense, or if any of that is correct! But its certainly making me think - which I'm glad about as DD will be outgrowing her (RF ) Grp1 seat soon and we need to decide what to get her next