The graph makes it pretty obvious that the effects isn't real. The newy steady growth in ratio 1910-1950 is more interest, but even so, the magnitude of variations is tiny.

If anything, infanticide is likely the cause of distortion. Look at China's One Child Policy.