That measure inherently excludes the “wealth” represented by ownership of enslaved people. It includes what those people produce, but the “per capita” measure also includes enslaved people in the denominator.

The production of enslaved people represented a small fraction of the economy—about 5% of GDP during the peak of the cotton boom: https://fee.org/articles/no-slavery-did-not-make-america-ric...; see also https://jacobinmag.com/2019/08/how-slavery-shaped-american-c...

More importantly, you can’t increase a country’s overall level of production by enslaving people. Its basic economics—unfree labor misallocates labor resources to industries controlled by slave owners. Slavery obviously redistributes wealth from enslaved people to a slaveholding aristocracy, but that comes at the expense of economic growth for the country overall. This is true both theoretically and empirically. As shown in the GDP chart linked above, the end of slavery didn’t produce an economic collapse, as you would expect of it really was responsible for America’s productivity. There was not even a dip. In fact, ending slavery made America richer, exactly as economic theory would predict: https://www.econlib.org/archives/2014/09/ending_slavery.html