Hacker News Reader

Comment by emiliobumachar | Hacker News Reader

I think the only way to avoid this brand of self-interest is to, prior to any discussion, agree that the changes will only come into effect 20 years into the future. It's hard to predict who it'll benefit that far into the future. 16 or 12 years might be doable. 8 is definitely too close. 4 is a train wreck, everyone is on campaign already.

View on HN

The downside of doing so is that it still gives plenty of time to cancel the changes at year 12 or 16, once it becomes apparent who will stand to benefit/lose from the change. This type of hacking future benefits is evidence in many of the tax overhauls, which use predicted future savings to offset short-term costs. And often those future savings are legislated away before they come to fruition.

(Of course, sometimes those future savings wouldn't have happened anyways, due to unreasonably optimistic assumptions about the performance of the economy.)

Cancelling such a scheduled change is harder though than simply blocking a change in the first place.

Exactly. Sunset provisions have large effects even though in principle they can always be removed if the majority in the future wanted to.

I encourage you to look into the history of copyright extension laws.

It's not that simple. First, I'm not sure that anyone knows who would benefit 8 years out. There will be guesses, sure, but nobody's really going to know how it plays out until after the first election under the new system. Even that won't really tell us much, because everyone will adjust their tactics after the first experience.

Second, even if everybody knew who was going to be the loser from the changes 8 years in advance, that's not enough. That party would have to be strong enough under the current system to undo the changes. In practice, that would require controlling the House, the Senate, and the Presidency... unless both parties decide that they would lose power to third parties, and strike a deal.

I was thinking they need to start recording and widely publicizing exit polls using the approval system so that people can get a sense of what the alternative 'feels' like. The big problem is that it'd take at least ~three election cycles for most people to appreciate the difference it'd make.

E.g., in the first cycle, extreme candidate A wins the actual election while the moderate, B, is shown to be widely tolerated under the approval system. Nearly half the voters--those who voted for the opposite extreme, C, end up loathing the results of the actual election and wishing in retrospect that the approval system had been in place. Perhaps in the second election cycle, candidate A wins re-election because they're a known-quantity/incumbent.

Finally, in the third cycle, history starts repeating itself and people get tired of A's party. E.g., the Overton window slides back towards C's perspectives leading to C being seated. This time the people who still prefer A are disgusted with the outcome and prefer the outcome of the approval system.

At this point, half the voters have long argued that the approval system makes more sense and the other half has also come around to thinking maybe it could bolster their position as well. It'd only remain for less partisan activists to e.g., get a petition going towards a referendum.

Or to (infuriatingly for the general public) simultaneously pass a package that ensures present-day personal benefits to all current incumbents, removing all risk from their calculations of personal benefit and thus inducing them to do what is simultaneously best for them and for the future polity.

> I think the only way to avoid this brand of self-interest is to, prior to any discussion, agree that the changes will only come into effect 20 years into the future. It's hard to predict who it'll benefit that far into the future.

It's hard to predict the short-term beneficiary parties when it goes into effect, sure, but its quite easy to predict the long term ideological benefit of increases in proportionality on modern democracies.

Ironically, that may make near immediate implementation more viable, if their is a divergence between the expected short-term and long-term impacts, and those expected to benefit believe they can use that to preempt the naturally expected long term benefit but those who would expect that long term benefit don't believe their opponents will succeed at that.

That can result in last minute scuttling too.

While politicians are notoriously bad at planning 20 years ahead, I wouldn't put it past them to try, especially when that timeframe makes it easier to project the impact of demographic trends rather than merely political ones.

Reply on news.ycombinator.com