Let’s back up. Let me know if I am mischaracterizing your views, Noctolator.
You believe each party should have a “core value” in order to identify it from its adversary and so that you, as an elector, know what you are buying when you pull the lever.
I contend that this is neither possible nor desirable. In a nutshell, your two questions are at the heart of political inquiry. I have a book that I think you might get something out of, but I doubt you will enjoy it.
The first problem, that parties should have some “stable” and differentiating core values is explained fairly neatly by one of the many flavors of median voter theory. To make a long story short, if voter preferences have a few formal characteristics (single-peakedness, among others) and can be arranged in a number line, the preferences of the median voter dictate.
Suppose 100 voters under simple majority electoral rules. Suppose two parties. Suppose one issue: the tax rate. Suppose your platform is equivalent to the tax rate you would set when you are president, and say this is a number between 1 and 10, with three decimal places.
Everyone has preferences over what tax rate he wants to be charged. Now, sort everyone by preference over tax rate from 1 to 10.
One party’s platform favors high taxes, the other party’s platform favors low taxes.
The utility of each voter is determined by the difference between his policy preference before the election and the actual tax rate implemented after the election. If I like tax rate 3 and that is the actual rate, I get buckets of utility. If I like rate 1 and we are stuck with rate 10, I am not so happy.
So the candidates are running for election. Both want to capture 51 votes. What tax rate do they choose?
They choose the tax rate of the median voter on the number line. By choosing a platform to maximize this voter’s utility, the parties are guaranteed the votes of everyone else within their class.
This is an excruciatingly simple version of the model. The actual outcome is that both parties converge to the same platform, no one votes (not even the median voter), and the election is determined by a coin toss. But the logic is crystal clear.
Parties will attempt to maximize the utility of the median voter. The reason republican and democrat platforms are not identical is largely because people are not necessarily single issue voters, so they have to deal with multi-issue space. This is a very difficult problem.
So the answer to your question is any party that tries to proceed consistently from a core of values rather than outcomes that maximizes the preferences of the median voter is doomed to irrelevance. The Libertarian Party, which likes to call itself “the party of principle” is case in point. Its platforms do proceed from relatively unchanging axioms.
It also wins few votes.
This is not an “unstable” system. In fact, it is remarkably stable and has been in a comfortable equilibrium for a long time.
The second problem is that voters never know what they are getting into when they pull the lever. This has nothing to do with core values but with a gigantic issue called “the credible commitment problem”.
Parties choose platforms to maximize vote share. Individual politicians seek office to maximize personal utility. Politicians make promises to voters that they are not externally compelled to keep. So the question becomes, how much can a politician promise and actually deliver to voters while still maximizing his personal utility? How much can he renege on his promises without getting thrown out of office?
A politician wants to deliver the bare minimum, and the electorate wants to elect someone whose preferences are concordant with theirs. But all preferences are private information, so everyone is constantly guessing.
The literature on this is enormous. Suffice to say, core values are not the answer, as they are promises as empty as anything else. There are few good answers. Most of them hinge on electoral rules and, even more so, institutions.