OK, so there are 2 main reasons for adding the boiling water to kill bacteria in the powder, and it's to do with different types of bacteria.
- the risk of general bacteria/spores from the non-sterile powder growing in the milk after preparation, including stuff like salmonella, etc., which do grow albeit slowly in the fridge (why you should discard after 24hrs, for example). This kind of bacterial growth might cause gastro-enteritis in a baby, which could be reasonably mild or quite severe, so it's worth preventing the risk of this as much as possible. However, as pp have said, in Western countries with clean/sterile equipment and good hygiene this is relatively low risk compared to the risks of bacterial contamination in other countries.
- Formula samples do and have tested positive for various types of more dangerous bacteria, such as variants of clostridium, including the one that causes infant botulism. This is v v rare (vanishingly small number of cases in the West), but can be fatal. It's why NHS advice is not to give honey to infants under 1 - because it can potentially contain C. botulinum, and infants' gastrointestinal tracts are too immature to have developed defences against this. Again, the likelihood is very very small, but the consequences could be fatal. The boiling water is meant to help kill any spores in the powder from rare but more dangerous bacteria such as this.
The guidelines were changed because several studies were done in the last decade or so on the composition of bacterial spores in baby formula and more widespread contamination by more dangerous bacteria was found in the powders than had been previously thought.
So:
When most posters are saying that they made up bottles the old way and all their children were fine and never had a tummy illness, that's because they were lucky that their babies (a) didn't end up getting gastric symptoms from more ordinary bacterial growth. But (b) they were also lucky that they never encountered any more dangerous bacteria either.
The risk of (a) is higher, but generally less serious in developed countries. The risk of (b) is statistically small, but potentially of much greater consequence. But guidelines were changed to reflect updated thinking on risk prevention, just as they were with the advice on honey.
Those who would be happy to make up bottles these days with non-boiling water: would you also be happy to feed your 6 month old honey, on the grounds that the risk is statistically very small? Or would you aim to avoid that risk just in case, because you can? I mean, people have been giving babies honey for centuries, and the vast majority didn't die of infant botulism, so why not? Or do you think of that differently because avoiding honey is a lot easier than boiling water for each bottle?
(US posters - I'm v surprised about the tap water thing, but then the US has much lower food safety standards than here; however US researchers have found species of Clostridium in formula there too....perhaps a case where the NHS is super-cautious, but then, why not?)