> Divorce rates would tend to decrease if marriage rates decrease.

I agree. That is happening. So the premise that divorce rates are ever increasing is incorrect as applied to the whole population and it is wrong as applied to the married population.