Mining stocks are the only group with a negative correlation to the rest of a portfolio over 10-year periods and very low correlation over 5 years.

Judged as of January 2016. TRUE: correlation data from GMO or an independent analysis show that the returns of mining stocks over rolling 10-year holding periods were negatively correlated with the rest of a diversified (equity) portfolio, that no other industry or sector group showed a negative 10-year correlation, and that the 5-year correlation was low (roughly below 0.3). MOSTLY TRUE: the negative 10-year and low 5-year correlations hold, but one minor group shows a comparable reading, or the 5-year correlation is only moderately low. MOSTLY FALSE: mining's long-horizon correlation is low but positive, or other groups are also negative so that 'the only group' is overstated, or no analysis can be found to support the claim beyond the speaker's own assertion. FALSE: the evidence shows mining stocks' 10-year correlation with the rest of the portfolio is clearly positive and not unusually low.

Set before the research began.

1. E1 Correlations for the long run - Outcast Beta Outcast Beta (Markku Kurtti) · 26 May 2024

   “Coal, Gold, Mines, and Oil industries have negative long-term correlations with the stock market.”

2. E2 Correlations for the long run - Outcast Beta Outcast Beta (Markku Kurtti) · 26 May 2024

   “Ships, Smoke, Agriculture, and Healthcare industries have long-term correlations with the stock market between 0 and 0.25.”

3. E3 An Investment Only a Mother Could Love: The Case for Natural Resource Equities GMO (Lucas White and Jeremy Grantham) · September 2016 Own record

   “the 10-year correlations have actually gone negative!”

4. E4 An Investment Only a Mother Could Love: The Case for Natural Resource Equities GMO (Lucas White and Jeremy Grantham) · September 2016 Own record

   “There aren’t enough non-overlapping 10-year periods for the 10-year correlations to be statistically significant, but we believe there is intuition for the”

An independent analysis [E1] finds that mines have negative long-term correlations with the stock market, which supports part of the claim. The same 2024 analysis also finds negative long-term correlations for oil, and for coal and gold, which it lists separately from mines [E1]; coal and gold may fall within 'mining stocks', but the oil reading still casts doubt on mining being 'the only group'. Other groups (ships, smoke, agriculture, healthcare) also show low correlations between 0 and 0.25 [E2]. GMO's own paper says its 10-year correlations went negative [E3], but also that they are not statistically significant [E4], and no evidence gives the 5-year figure. Under the criteria, 'the only group' is overstated, so the claim resolves mostly false.