Firstly, there's the role of ideology. If the evidence conflicts with someone's ideology, they'll continue to support policies that run contrary to the evidence. On cannabis, those who think taking any recreational drug is a stain on society will support policies that ban drugs, even if the evidence suggests that cannabis criminalisation does no good. If mass surveillance did a lot of good, and prevented an incredible number of terrorist attacks every year, then those who care about liberty in the abstract would still oppose it, even if they currently argue that mass surveillance does no good anyway.
Eliezer Yudkowsky wrote an
excellent piece a while back, titled "Is That Your True Rejection?". As he wrote:
Thus, the fact that people have fundamental moral disagreements with each other which can, at best, be resolved by rational argument and delving into moral philosophy, is the first reason that people listen to politicians (that they agree with) over the evidence.
Secondly, people don't have time to research evidence-based policy. (EDIT: Well, they do, but they
don't care. People spend all their time watching reality TV shows, buying things they don't need at shopping malls, and sleeping). They'll rely on the science reporting in newspapers, which is very often atrocious and/or biased, as John Oliver covered recently on his show, Last Week Tonight.
I would say that those are the top two reasons: the usual suspect - cognitive biases such as confirmation bias - as well as practical difficulties such as time.
You can resolve disagreements between experts, and probe their arguments to discover who is right or wrong, though. You're essentially arguing, here, that there's no right answer to factual, empirical questions, but there is.
The OP is saying that the politicians are getting it wrong on factual, empirical questions that are answerable, not value-judgements.
Let's take the sugar tax. Medical doctors are qualified to say whether we should implement a sugar tax. Their role is to improve the health of the population. Public health officials' jobs are to improve the health of the general population. I agree that politicians should make policy decisions based on the synthesis of all of the experts' views, but the OP is saying that politicians don't make evidence-based policy decisions, and that they are allowed to do so by the electorate, who actually believe politicians instead of looking at what the experts have to say, and he's asking: why?