Comment by 3ygun | Hacker News Reader

Comment by 3ygun | Hacker News Reader

Parent Full thread

As the parent notes. The issue is if you have a simple law say "Do not dump harmful chemicals into waterways used for drinking, shipping, etc." this leads to a lot of litigation regarding what is a "harmful chemical". If instead you have it all spelled out "harmful chemicals being one of coal ash, hydrochloric acid, asbestos, ..." (don't quote me on these). Business's know what the variables/costs effecting operations are up front. Only reacting and being liable if the list changes through the political process.

The parent was saying the US approach instead define what a harmful chemical is after the fact. Leaving companies to need "insurance"/a cost benefit wager before something is built. Otherwise they could be screwed later after litigation.

IMO, I can see both sides but preferably I would want simpler laws from political processes which are then refined into explicit interpretations by subject matter expert strong government institutions. But we have lobbying/revolving door government employees and trashy textualists in the US so it doesn't work out that way...

[1] https://en.wikipedia.org/wiki/Textualism

I think the modern US version is different:

A federal 3 letter agency unilaterally writes detailed regulations, that are for practical purposes law.

One problem is that the regulatory agency gets captured by the industry, which means the regulation ends up being a way for the established industries and companies to make competition unfeasible.

We'll see how long that continues to be the case. Chevron deference is under fire from our whackjob supreme court.

https://en.wikipedia.org/wiki/Chevron_U.S.A.,_Inc._v._Natura....

The regulatory capture conversation from the likes of pmarca is always about advocating reduction of regulation because otherwise the regulator starts working in favor of the big guys against the little guys. But to me that is the wrong way to look at regulatory capture. The antiregulation politicians are the ones that tend to put big business lobbyists in charge of regulation. We need good regulation, not less. Good regulation that evens the competitive playing field and protects exploitation of externalities. without regulation the bug guys lock in the rents with no space for competition. We need good regulation. But you don't vote for the antiregulation guy to get good regulation, he's trying to destroy regulation not improve it. Less government is a red herring, we need better government.

Thing is, good regulation is extremely tough. A good regulator needs to deeply, deeply understand the industry, where it has been, where it is going, and what the problems are; they need to deeply understand the spirit of existing laws rather than the letter of the law and be ready to thoughtfully and creatively deal with edge cases; they need to have some risk tolerance to allow for growth while still protecting people; they need to be able to balance the interests of competing groups; they need to be able to be proactive / innovate rapidly in the face of changing conditions meaning they need to understand the bleeding edge and be consulting with the groups at it; they need to see big problems like pandemics coming and push people in the right direction; they need to resist the calls for deregulation for deregulation's sake from industry as well as resist the calls for regulation for regulation's sake from politicians who want to appear like they're doing something about a problem; they need to deeply care about the users they are charged with protecting; Finally, they have to have brass balls. They have to be unafraid of standing up to politicians / the public / industry and saying you are wrong or sorry, we were wrong.

It's really tough to find lots of people who can do all that for not much money / acclaim potential.

I believe what you call pmarca's view is also the view of researchers in the field. The term "regulatory capture" comes out of the Public Choice branch of Economics.

> We need good regulation, not less

The result of the research is that that is, in general impossible. Sure, some regulators do good work for some period. But the general forces and incentives in play push quite strongly towards the regulator being captured.

The better working alternatives are laws, where disputes or guilt are decided by courts. Since there is no regulator to capture in this scenario, things work pretty well. Of course, lawmakers can also be captured, but that's harder.

One fascinating* thing with the field of economics is that so m any thinks their opinion is worth as much as the expertise of professional in the field.

You routinely hear people confidently claim the economics field is wrong about things they've studied for decades or centuries, because of simple arguments that are all thoroughly accounted for in the beginner literature.

* or "infuriating", if you are an economist :)

Sorry but an argument from authority in the field of economics is not even a little bit convincing to me. Not only is economics a "social science", which are notorious for lack of rigour, but economics itself is purposefully bifurcated into biased partisan "schools". There are many economists which study solutions to regulatory capture who do not suggest that regulators should be cut, if you're only aware of one school of thought in this area then you've been looking in a limited bubble.

Not sure that is how the cookie crumbles, rather: it will be decided that chemical X is harmful at a certain dose, which leads to some permissible concentration Y of X, and that Y can then be used for regulation of disposing X.

A place that needs to dispose a certain volume of X with concentration >Y then pays a sort of a mixologist to dilute X to a concentration <=Y before dumping it all anyway (unless it would be less expensive and more practical to react X away).

This is a great description. And, to be clear, I"m not really advocating for either system being better. Just pointing out some of their differences.

Reply on news.ycombinator.com