Of course "overweight" has as-good or better long term health outcomes so... Meh? https://academic.oup.com/ije/article/35/1/55/849914
That said, the section on health outcomes is not hugely compelling. The "normal weight has worse outcomes" studies (particularly the famous NEJM one) have a significant set of problems.
NEJM simply didn't use the same bucketing as common parlance, so bucketed comparisons (as opposed to BMI ones) are basically meaningless.
More broadly, these studies are generally correlational, which produces serious issues regarding patients suffering from diseases. Eating disorders have very high mortality and often cause low weights, while conditions like cancer and neurodegenerative disease often cause both mortality and low weight. Without correcting for that, you're actually noticing trends like "late stage cancer patients are both skinny and sick".
I'll have to go and dig up references, but I thought the current state of research suggested that among otherwise-healthy patients, especially men in middle age or older, health outcomes noticeably worsened beyond the middle of the overweight category?