In spite of this, they are routinely ignored. I suspect this is because they don't produce the same human drama we get from a poll, they are morally suspect because they take very emotional questions of meaning (e.g., "who are we as Americans/Scots/whoever?" and turn them into relatively dry statistical calculations, and betting is already morally suspect among a lot of people. We're poorer for it: we're substituting noise for signal by talking about polls when we should be talking about betting markets.
1. Identify a few strong candidate policies that are a) widely endorsed by economists, b) based on relatively simple clean analysis, c) not much adopted in the wider world, and d) should bring big gains.
2. Try to engage other intellectuals in detail on one or a few of these, seeking to either gain their endorsement, or to understand better the barriers that block them. If possible, do this as a group, and using all our status levers to make them respond in detail. If we succeed in persuading intellectuals, then join with them to try to persuade policy-makers, again either succeeding or better understanding barriers.
3. Once we better understand barriers, focus our economic research on doing what it takes to overcome them.
I think betting markets likely fit #1. So what's the next step?