For many couples considering parenthood, the most frightening risk is the possibility of a child with a serious birth defect. That is scary, but also unlikely, especially for couples willing to use amniocentesis to check for serious problems and if necessary abort. The risk that is both frightening and reasonably likely is the chance of having a child who does not like you.
I got along well with my parents, better than I did with most other people I knew, my wife got along well with her parents, and our children get along well with us. But my impression, largely from our children's accounts of people they know, is that a large fraction of teenagers and young adults, perhaps a majority, do not get along with their parents. That not only makes the process of bringing up children a lot harder and more unpleasant, it also eliminates one of the major long-term benefits—ending up with adults whom you love and trust and who love and trust you.
Which raises the question of whether we have just been lucky and if so in what way, or whether we, and our parents, did something right. Judith Harris' work, presented in her very interesting The Nurture Assumption, provides evidence that parental child rearing does not have a very large effect on the personality of the child when he becomes an adult, but it might have a large effect on the adult's relation with his parents. And even if child-rearing does not have a large effect, other environmental influences, especially, by her account, the peer group environment, do.
Harris mentions that in some families, although not many, the family is the peer group, creating an exception to her general rule. That, I think, describes my situation, my wife's situation, and the situation of our children. In each case, the child identified more strongly with the family culture than with the culture of his age peers. It might also describe a situation more common when population was much less dense, with the result that a smaller fraction of social interaction occurred outside the family. For an extreme version, consider the situation of the (fictional) Swiss Family Robinson. For a less extreme one, consider any family that is committed to a different view of the world than the surrounding culture—conservative Christians in a secular environment, atheists in a religious one, or immigrants from a very different society. It would be interesting to know whether such situations results in significantly less parent/child conflict.
Or maybe it is pure chance. There seems evidence that personality is to a significant degree genetic. There may be personality types that do not get along with each other, even types that do not get along with anyone. If that is the fundamental problem, we have been very lucky.