When electing a leader we look for two things. First we want someone who will pass the laws we want them to pass (universal healthcare). And secondly we want a competent leader, someone who is able to make good decisions, compromise with others when necessary, and react well in a crisis. Current implementations of democracy roll these two into one; when it comes time to vote you have to pick a representative who is going to be a good leader and who will implement the policies you want. When you think about it, that is a really stupid way to do things. By doing things this way we often end up voting for poor leaders just because they support the policies we like. And because our choices in public policy are tied to candidates there is often no candidate who supports everything we want; often candidates pick the issues they back so as to divide the voters among them, which makes it harder for candidates representing compromises between extremes to win.
There is no reason things have to work this way. It is perfectly possible to vote for the leader who we judge to be the most competent and then, in a second election, vote on the issues. Afterwards we would expect the leader we had elected to act according to the expressed wishes of the people, as expressed by the vote on the issues. Obviously there is the possibility that the elected leader won’t respect the will of the people and instead act as they want rather than as the people want. But this is not a new danger, the possibility already exists in the current system for candidate to betray the trust of the people, say by acting contrary to the platform they campaigned on. As things are currently done we suppose that such behavior can be discouraged by not reelecting anyone who acts that way, and similarly in the alternate system we could refuse to reelect anyone who acted contrary to the expressed will of the people.
One improvement that would be brought about by such a system is that it would virtually destroy political parties. Although nothing could prevent a political party from forming they really wouldn’t have any role to play. If such parties defined themselves by their positions on various issues then no candidate could afford to belong to them, since it would throw doubt on whether they would really listen to the people. And, if they followed the party’s wishes rather than the people’s wishes, well then they wouldn’t be elected again. Political parties might survive by reinventing themselves as a kind of standard of credibility, meaning that being endorsed by those groups would indicate that the candidate is well qualified to be a leader. Such parties would be basically harmless, unlike political parties as they currently are.
A second improvement that would be brought about by this system is that it would motivate elected officials to create compromises that satisfy people on both sides of the issue. Suppose that on a certain issue the populace is basically evenly divided. Now the official in charge could choose to obey the will by just doing whatever was slightly more popular, but that would leave a large number of people feeling dissatisfied, which might make reelection hard. But if the official devises a compromise that makes both sides reasonably happy (or at least keeps them from being too unhappy) then people on both sides of the issue will be willing to reelect them (and the compromise itself will probably make for good PR when it comes to election time, as compared to a decisive decision for one side or the other, which will make almost half the population unhappy to be reminded of).
And, most importantly, this system gives elected officials the freedom to do what they think is best, even if that is contrary to the will of the people. They have this freedom in virtue of the fact they are elected because of their competence, not because of their position on specific issues. So when it comes time for reelection an official can justify their correct, but unpopular, choices by pointing out how well things turned out. Similarly an official might lose their position as a result of making a popular choice that turned out poorly. Their opponents would simply have to point out that by just doing what was popular they have proven themselves to be an incompetent leader, because the best leaders make unpopular choices when they know the popular choice is wrong.
So, to summarize, by voting for officials and on issues separately we can avoid the many problems that arise from politicians and political parties gaming the system by selecting their positions on issues so as to divide the voters to their benefit. Because people tend to care more about the issues than the candidate (perhaps justifiably) it is also easy under the standard system for incompetent leaders to be elected. And because candidates and issues come as a package deal it is often the case that no candidate represents the truly democratic outcome. But, despite the inferiority of the standard system to this alternate system, it is practically impossible to switch from one to the other. Since people are so used to voting for candidates in order to express their opinion on issues even after making the switch candidates would still take stands on various issues and people would still vote for them based on their positions. Which wouldn’t work if we had started originally with the alternate system, since a candidate taking stands on issues would be effectively announcing that he or she intended to ignore the will of the people, and would thus be unelectable. (Or, to look at it from a different angle, say they took a stand on an issue. If they take the minority position they won’t be elected. But if they take the majority position there is still no reason to elect them on that basis, since their opponents will also take that position once they are elected, if it really is the majority position, and so the majority has no reason to vote for them.)