In many social and economic problems, sometimes it is quite clear "what" has to be done. It is less clear "how" to do it. Also for more personal problems: it is easy to advise an obese person to eat less and exercise, it is more difficult to tell him "how" to manage to achieve it in a sustained way. Three (related) examples come to my mind:
-Many claim that some decisions should be left to experts, and not to amateur politicians. However, it is less clear how in a democratic society, the majority will delegate their decision making powers on élites that may not share their distributive preferences.
-Competition and merit should play a larger role in the allocation of resources. However, there are losers from competitive processes, and the losers will make their voices be heard in a democratic polity.
-The selection of politicians should be focused on looking for the best citizens and the reform of political parties should focus on improving the quality of policy-makers. However, politics is labour-intensive and some citizens invest their human capital in this career, and will not easily accept to be replaced by part time well trained elites.
Perhaps, as Bowles reminds us in his book "Microeconomics", the problem is that any efficiency-enhancing reform is accompanied by distributional changes. In European countries, it is very common to suggest (allegedly) efficiency-enhancing reforms without proposing ways to alleviate the distributive tensions that accompany them.