Robin Leichenko published an analysis for the 1970s, 1980s and 1990s and also found a complex pattern but leaned to complementarity.
I had previously mentioned that Wendell Cox pointed me to population data that go back to the 1950s. A sample of twenty-four areas across the U.S. is available if we want populations for the urbanized areas, their core cities and the remaining suburbs.
What were the inter-decadel population growth correlations?
The positives have it, but the associations get weaker over time. Growth prompts now come from far and wide.