I suspect there are lots of different answers to this interesting question, but they all come down to which strategies are most successful in producing offspring that survive to adulthood.

One way of doing that is to provide parental care for a small number of offspring, to ward of predators; another is producing lots of offspring an hoping some survive on their own. Each works well under different circumstances.

Once parental care is established in a lineage, there are advantages to having more of the development happen after birth or hatching. For example, less energy is needed to be invested by the mother carrying them. The mother is also less vulnerable to predators if birth happens earlier, so there is a balance between the survival of the offspring and the survival of the parents. In social organisms, where the parents are not the only adults to care for the offspring, obligate care is obviously a better survival strategy for the young.

It's interesting how vertebrate parental seems to be much almost ubiquitous in warm-blooded groups (mammals, birds) and virtually absent in cold-blooded groups (except crocodilians). I wonder whether it's associated with the need for incubation (or internal development) to keep them warm until hatching/birth, which is behaviour that is easily extended into extended care..?

I think that parental care would amost always be necessary in mammals, as the young in the placenta need to be comparatively very small, and are initially dependant on their mother's milk, and probably also depndent on their parents strength to defend them. Many vertebrates lay eggs, and parental care is not absolutely needed, but is often rather essential, and given, in the form of bringing food to the young. Some birds lay their eggs in the nests of other bird species, tricking these non-parents to give the parental care of bringing food to their young, etc. The non-parents never realize that their own young ones have been pushed out of the nest and left to die by these intruders . The difference between warn blooded groups and cold blooded groups oulined in the previous answer is also very relevant. And then there are the insects and caterpillars and butterflies. It is all so very complex and wonderful.

Smart species take a long time to grow up and recieve more parental investment. Being smart is one of the best ways to adapt to an environment and it carries the advantage of not being locked into instictual and anatomical restraintsa when the environment changes. Behavioral plasticity is depoendent on parental investment.

One of the interesting outcomes of the evolution human cooperative breeding is that it has made human mothers capable of producing a much greater number of offspring than other great ape mothers. If we may assume for the sake of hypothesis testing that hominid ancestors represented a more ape-like pattern of long weaning and relative independence of young at weaning, at some point this shifts to shorter weaning, shorter inter-birth intervals, and a longer juvenile period of at least partial dependence. None of these behavioral shifts leave readily identifiable signatures in the paleoanthropological record of human skeletal remains nor in the paleolithic archaeological record. There are some suggestions that faster development of offspring may be apparent from tooth eruption and epiphyseal plate fusion in juveniles, however our sample of infant and juvenile hominins is very small because of preservation problems with small bones and no apparent burial of the dead until the time of the Neanderthals. The increasingly long juvenile dependence period can be seen as potentially under selection as useful in "educating" offspring about the complexities of modern human subsistence and social practices, but it also sets up a dilemma regarding how mothers care for dependent offspring of several different ages and requirements for food, protection, education, etc. Popular ideas have focused on male provisioning or the "grandmother" hypothesis, however both of these positions would require that the evolution of shorter initiation of weaning by human mothers, and at least some of the biological and reproductive changes changes have already occurred. There would be no selective advantage to investing in offspring by fathers or grandmothers at the levels seen in modern people, unless mothers could not accomplish all of that by themselves (as they do among other great apes). Additionally, at least some aspects of the sexual division of labor and probably some for of monogamous mating (even if only serial monogamy rather than life-long pair-bonding) would need to have developed for any selective pressure on males, grandmothers, or other relatives to assist in provisioning young. It seems much more parsimonious to address the changes from a more ape-like pattern to one more like modern humans in order to establish the direct and indirect selection benefits that can be realized in the kinds of extensive cooperation and sharing seen among modern people. See the following modeling paper by Karen Kramer & Erik Otárola-Castillo (2015) and other work by Kramer and other collaborators that do address ways to model potential changes in mother/offsping relationships in order to establish some of the reproductive behavioral changes that make humans realize much higher fitness than out other great ape relatives through cooperative breeding:

Kramer, KL & E Otárola-Castillo (2015) When mothers need others. Life history transitions associated with the evolution of cooperative breeding.

Journal of Human Evolution 84:16-24. Published, 05/2015. http://www.sciencedirect.com/science/article/pii/S0047248415000196

Kramer, KL & AF Russell (2015) Was monogamy a key step on the hominin road? Reevaluation of the Monogamy Hypothesis. Evolutionary Anthropology 24:73-83. Published, 03/2015. http://onlinelibrary.wiley.com/doi/10.1002/evan.21445/epdf

Kramer, KL, R Schacht & A Bell (2017) Adult sex ratios & partner scarcity among hunter-gatherers: implications for dispersal patterns and the evolution of human sociality. Philosophical Transactions of the Royal Society, B 372 20160316 http://dx.doi.org/10.1098/rstb.2016.0316. Published, 07/2017.

Kramer, KL (2015) Cooperative breeding & human evolution. In Emerging Trends in the Social and Behavioral Sciences, RA Scott & SM Kosslyn (eds). John Wiley and Sons, Hoboken, NJ. DOI:10.1002/9781118900772.etrds0055. Published, 01/01/2015. http://onlinelibrary.wiley.com/doi/10.1002/9781118900772.etrds0055/abstract

Kramer, KL and AF Russell (2014) Kin-selected cooperation without lifetime monogamy: human insights and animal implications. Trends in Ecology & Evolution 29(11):600-606. Published, 11/01/2014. http://www.sciencedirect.com/science/article/pii/S0169534714001931

Kramer, KL (2014) Why what juveniles do matters in the evolution of cooperative breeding. Human Nature, special edition on eusociality, 25(1):49-65. Published, 06/2013. http://link.springer.com/article/10.1007/s12110-013-9189-5#page-1

Thank You For Your Interesting Answers

I fully agree with most answers above, but in the case of H.sapiens there might be a more specific reason why we need longer puberty, and that is the brain enlargemnt we see in the genus Homo, and which was arguably largely due to the intercontinental dispersal of early-Pleistocene Homo along African & Eurasian coasts & rivers, where brain-specific nutrients (DHA, iodine, taurine & oligo-elments) were abundant, esp. in shellfish ("seafood is brainfood"), google e.g. "coastal dispersal of Pleistocene Homo 2018 biology vs anthropocentrism". After the littoral phase, whne H.sapiens went inland (when DHA etc. was scarcer), they might have needed more time (longer puberty) & more support of family members to acquire sufficient nutrients for their large brains?