This is untrue. If it really were a universal, why would there be such high IQs in the first place? As well, it's not empirically obvious: some samples show dysgenics, others show eugenics. I'm sure you're familiar with Clark's claims about England and China, but you can also find eugenic samples in a number of modern samples (iirc, post-WWII Baby Boom was eugenic) collated in Richard Lynn and Tatu Vanhanen, Intelligence: A Unifying Construct for the Social Sciences, 2012; and some HBDers crunching the GSS claim that current trends are eugenic for men but dysgenic for women.

What about the reaction time series that suggest a consistent decline in g?

Deeply questionable, since there's no real way to check that the populations remained the same by SES or ethnicity, and the apparatus themselves are not necessarily identical: a recent comment on the topic I read observes that the entire small decline could be explained by shifts in the luminance of the devices (the brighter the light, the faster you can react).

This is true - otherwise we'd have consistently growing intelligence over the generations - and instead we find the inverse - dysgenic fertility rules the day.