The Grandjean (China) studies could be not replicated in other countries using a lower standard of water fluoridation, particularly in New Zealand, which uses a similar level of fluoridation as the US. Quoting relevant bits of the 2015 PHS recommendation on this matter:
IQ and other neurological effects. The standard letters and approximately 100 unique responses expressed concern about fluoride's impact on the brain, specifically citing lower IQ in children. Several Chinese studies considered in detail by the NRC review reported lower IQ among children exposed to fluoride in drinking water at mean concentrations of 2.5–4.1 mg/L—several times higher than concentrations recommended for community water fluoridation.[81–83] The NRC found that “the significance of these Chinese studies is uncertain” because important procedural details were omitted, but also stated that findings warranted additional research on the effects of fluoride on intelligence.
Based on animal studies, the NRC committee speculated about potential mechanisms for nervous system changes and called for more research “to clarify the effect of fluoride on brain chemistry and function.” These recommendations should be considered in the context of the NRC review, which limited its conclusions regarding adverse effects to water fluoride concentrations of 2–4 mg/L and did “not address the lower exposures commonly experienced by most U.S. citizens.” A recent meta-analysis of studies conducted in rural China, including those considered by the NRC report, identified an association between high fluoride exposure (i.e., drinking water concentrations ranging up to 11.5 mg/L) and lower IQ scores; study authors noted the low quality of included studies and the inability to rule out other explanations. A subsequent review cited this meta-analysis to support its identification of “raised fluoride concentrations” in drinking water as a developmental neurotoxicant.
A review by SCHER also considered the neurotoxicity of fluoride in water and determined that there was not enough evidence from well-controlled studies to conclude if fluoride in drinking water at concentrations used for community fluoridation might impair the IQ of children. The review also noted that “a biological plausibility for the link between fluoridated water and IQ has not been established.” Findings of a recent prospective study of a birth cohort in New Zealand did not support an association between fluoride exposure, including residence in an area with fluoridated water during early childhood, and IQ measured repeatedly during childhood and at age 38 years.
The Grandjean papers are refs 84 and 85, above; the latter one is with Landrigan, in Lancet Neurology 2014. As you can see above the PHS opinion on it is that it is overstating its case, particularly as it applies to other countries.
And from a 2018 paper
However, some debate remains about the health
risks of CWF [community water
fluoridation]. For example, in 2012, a systematic review and
meta-analysis was published to assess developmental fluoride
neurotoxicity (5). The authors suggested that exposure to
“high” levels of fluoride may affect children’s neurodevelopment.
Reversely, a prospective study in New Zealand
published in 2015 which was conducted to examine the possible
correlation of CWF and intelligence, found that, after following
up participants between 1972 and 2012, there was no
significant association between fluoride exposure due to
CWF and intelligence quotient (IQ) (6). Some researchers
maintain there is not enough evidence to conclude that fluoride
in drinking water may impair IQ (7).
Yet, the digitization of mass media communications has
led to the far-reaching spread of unrestricted and inaccurate
fluoridation information across the web through media and
social networks (8). For example, based on the 2012 Choi
et al. article (5), an article was released in 2014 by one of the
coauthors. According to their review, the authors listed fluoride
as one of six newly identified developmental toxins in children (9). Several limitations in this article were addressed
by the scientific community (10,11); however, its publication
immediately triggered adverse coverage for fluoridation in
the popular press, generating tens of thousands of views and
shares over social media within 48 hours of publication, disseminating
flawed messages about fluoridation (12). The
safety and benefits of CWF continue to be debated online,
with reference to the 2014 publication, despite robust scientific
evidence that CWF is safe, protective, and effective (1-4).
In this latter paper, ref 5 is 84 from PHS (Choi,.., Grandjean) and 9 from the latter quote corresponds to 85 from PHS (Grandjean and Landrigan). And if you want to read the direct rebuttals of Grandjean and Landrigan (refs 10 and 11 above) they're freely available on Lancet's website: here and here which reiterate the same concerns. Quoting from the latter letter:
Grandjean and Landrigan did not acknowledge the animal study that showed no evidence of a neurotoxic effect of fluoride, even at levels up to 230 times the recommended concentration; an earlier study showing that fluoride causes no harm to children; two formal reviews that delineate weaknesses in the Chinese fluoride and IQ studies;[5, 6] and the conclusion by one of these sets of investigators that biological plausibility for a link between fluoridated water and IQ has not been established.
And from the former letter, which has broader criticism of how Grandjean and Landrigan put together their list of "neurotoxicants":
In their Review of neurological toxicants, Grandjean and Landrigan make unsubstantiated claims and misquote previous studies to pull together heterogeneous elements and drugs into a group of substances termed neurotoxicants.
[...] Their “strong evidence” for adding fluoride was a finding from Grandjean's own review of older Chinese studies. [...] By contrast with this review of Chinese studies, all of problematic methodological robustness, more than 3000 studies of the safety of water fluoridation stretch over 65 years. During this time, as fluoridation increased from 0% to 72% of US households, average US IQs have not decreased, but have instead increased by 15 points.
The investigators also added manganese to their list of neurotoxicants writing that “might cause” or has been “linked to” neurological disorders: prenatal exposure to ADHD and postnatal exposure to parkinsonism. However, recent reviews have shown no link between manganese and ADHD or parkinsonism.[6, 7]
Their assumption that “neurotoxicants might lurk undiscovered” behind even low doses of all chemicals can't be disproved. But it denies dose-response concepts. By seeing manganese and fluoride alongside repeated analogies to lead, a reader would naturally associate these elements with the situation with lead, in which no safe level of exposure exists. Manganese, however, as a trace element, does have a safe level.
(For meaning of "trace elemement" in this context see Mineral (nutrient) on Wikipedia)