You clearly have a lot of misconceptions about evolution. First, the human species has been more or less stable for at least 20 times longer than recorded history, so in evolutionary terms the "success" or lack of same of China or the US or any other "society" other than hunter-gatherers is pretty much irrelevant. In another 100,000 years humans may be showing adaptations to such societies but there is no evidence to speak of that that is currently the case.
Second, "success" in evolutionary terms has nothing to do with wealth or power or money or what kind of house you live in or anything else of that nature--the only criterion is reproduction--that genetice makeup that produces the most offspring who grow to reproductive age is the one that is most successful.
In some species a genetic makeup that leads to cooperation results in the largest number of mature offspring, in others it doesn't. Cats (leaving aside lions and hyenas) don't cooperate over anything but making more cats and even there a certain amount of violence is often attendant and yet they seem to do quite well.
So, taking your examples, first, China may not be "successful" in material terms, but more than 1/6 of the population of the world is in China so in evolutionary terms they are doing something right. In evolutionary terms those "developed" societies that you hold up as shining examples are _not_ doing well--technological development seems to bring about reduced population growth so they may very well turn out to be evolutionary dead-ends.
Now, as for your contention that there is a "bigger pie overall" in an "honest society", are you saying that an "honest society" will have more people of reproductive age than will a "dishonest society", because in evolutionary terms that is the only "pie" that matters. A homeless man who manages somehow to charm a large number of women onto his heating grate and manages to impregnate a small percentage of them will be more "successful" in evolutionary terms than will Bill Gates.
You say "Kerala is very poor in economic terms, but a very advanced society: low birthrate, high literacy, low crime, etc."
Read that carefully: "low birthrate". In other words it is eliminating itself from the gene pool.
You say "The better question, which offspring have greater success? The ones raised by a pair of loving parents or the ones with an absent parent?"
Which are going to have more kids, the impoverished children of broken homes who start reproducing in their teens and have numerous unplanned pregnancies or the ones who start in their 20s and have one or two planned kids? "Success" in financial terms is irrelevant.
You say "Hmm.. not sure what you mean. Are you saying its impossible to engineer machines that can reproduce and compete for limited resources? I doubt that its really impossible, but I wonder what purpose such machines would serve."
It is certainly possible to design such machines. But that has nothing to do with evolution. Their design would remain the same for all eternity if there was no means by which it could be altered. In "natural" organisms that mechanism is mutation and random chance--if a mutation results in more offspring growing to maturity then it stays in the gene pool, if it results in less then it eventually gets blocked out. Robots don't have a mechanism for mutation--they're designed to do what they do. If their design is changed it is because someone or some thing sat down with a CAD program or pencil and paper or whatever and changed it, not because a cosmic ray hit a DNA strand.