Is corruption a valuable antidote to overregulation?

post by ChristianKl · 2020-11-08T12:15:35.819Z · LW · GW · 8 comments

This is a question post.

Contents

  Answers
    16 betulaster
    13 Ericf
    3 George
None
8 comments

When reading Robert Caro's book about Robert Moses there's a part of my that things that I'm glad that the city in which I'm living is less corrupt then New York was when Robert Moses was active. 

On the other hand Robert Moses got things done and build a lot of infrastructure. After the Great Stagnation we aren't succeeding at building new infrastructure like Moses did.

In the discussion about why we don't have flying cars there's discussion about how nuclear plants in the US grew in cost while South Korea managed to continue to build cheaper nuclear plants. I read a few more recent articles and South Korea and it seems that after scandals that uncovered that a lot of corruption was involved in the nuclear industry when it managed to provide cheap nuclear power, and uncovering that corruption blew up their nuclear industry.

Corruption prevents bureaucracy from strangling progress as it gives people in power a way to get things done despite of the bureaucracy. 

While I have strong instincts that corruption is evil, it seems we lack a good discussion of it in relation to the evil of bureaucracy that strangles progress.

What are your thoughts on the issue?

Answers

answer by betulaster · 2020-11-09T09:55:08.508Z · LW(p) · GW(p)

There is a Russian saying (it's frequently ascribed to Saltykov-Shchedrin, and sounds to me like something he'd write, but I'm not able to properly source this) to precisely that effect... but not in the way you imagine. Roughly translated, it goes like this - "In Russia, the severity of the laws is compensated by the non-neccesity to obey them". 

(I don't have too much time for this, so apologies for shoddy sources down in the answer. Please let me know if you'd like more proper ones, I'll be sure to come back to that later.)

My personal model of Russian corruption maps very well onto Banfield's amoral familism. In a nutshell, the model is that every single individual is morally incentivized to increase utility of his small social ingroup (usually his family, but may be also a group of otherwise associated friends - part of Vladimir Putin's inner circle can be traced back to a housing co-op in the 90s), to the detriment of both his own and society's total utility. 

One addition from my own anecdotal experience, that I think Banfield never observed or made, is that an amorally familistic society needs a tyrannical/absolutist ruler to oversee it, lets it collapses into a hobbesian war of all against all - or, at least, that is the usual Russian perspective of it IMO. The tyrant sets the boundaries that make sure the ingroups don't slaughter each other completely in the struggle for utility, and enforces them with terrible, ruthless power. He is, in that sense, elevated from the earthly struggle for utility by that duty - that's why it's not really possible to blame him for any social woes, and instead the blame falls on those that carry out his decisions (and are "earthly" and subject to familism like all others) - there is another, IMO popular, saying to that effect, "the tsar is good, the boyars are bad".

(Ostensibly, this contradicts the saying from the very first paragraph - wasn't it possible to get around draconian laws by corruption? Yes it was. The idea is that while the tsar has his own, directly-controlled enforcers - the oprichnina or the rosgvardiya - and god help you if you cross the law and they notice, he also has a vastly more expansive hierarchy of appointed administrators and law enforcers (whom the average member of society has vastly more chances of encountering in his affairs), who are much more human, hence amorally familistic, hence corruptible.)

Regarding how this maps onto progress, it would probably be useful to consider Acemoglu/Robinson's "Why Nations Fail", keeping in mind two things.
First, Russia is not a progressive country by any means, empirically. 
Second, the idea that economic growth equals broadly understood progress is IMO defensible, but not very trivial. But since Acemoglu and Robinson are insititutionalist growth economists, who see technological/scientific progress as the driver of economic growth, it's still not useless.

The WNF take on progress is that people are incentivized to create/adopt disruptive technologies that drive progress only if they have faith in the social/political institutions that will ensure they can extract utility (for themselves, or whomever they care about) from that. If they instead know that institutions care about estabilishing the pre-existing mono/oligopolies, any disruption will be quashed, and progress will be sporadic and unsustainable in the long run. 

The Russian case appears to me to be almost that - the ruler is more concerned with maintaining social order than with maintaining the economic status quo (although those two may be connected), but since any large economic entity (like a large corporation running on a technology that is about to get disrupted) is by default more powerful than it's disruptive challenger, it works out the same. The amoral familism begets a ruler concerned with enforcing boundaries and the status quo, and disruption just goes agains the grain of both.

To conclude, yes, corruption does get used as a way out of overregulation, but not neccessarily to the benefit of long term progress.

(Acemoglu and Robinson also make a more direct connection by claiming that you neccessarily need a liberal democracy to estabilish these institutions that are conducive to progress and disruption. I'm hesitant about accepting that head on, but if you choose to agree with that, the case is even easier - amoral familism sees democracy as just ridiculous, and liberalism as weak. If I'm amorally familistic, I only care about satisfying desires of my ingroup - why would I ever want to live in a society which equates my desires to ones from outgroups instead? And if I choose to have a ruler that enforces social order and prevents an all-out collapse, how would he be able to enforce it with terrible punishments if he's limited by some "freedoms"?)

P.S. I have to note that Acemoglu and Robinson are concerned with long-term sustainable growth only. In the short term, however, it would probably be a lot easier for the ruler, or for currently endowed players (like oligopolies/oligarchs) to get things done by getting around red tape. 

answer by Ericf · 2020-11-09T02:33:49.317Z · LW(p) · GW(p)
  1. Attempting to distinguish between stifling regulation and ordinary "prevent defections" regulation is an inherently political process.
  2. Bypassing regulations is very rarely progress
  3. Progress has both winners and losers. Often, the difference between progress and regression is only visible with hindsight (eg leaded gasoline)

So, the key question is not "is X a viable antidote to overregularion" but rather "who gets to decide when a regulation has overstayed its welcome. And any process that doesn't include the voices of those currently benefiting from the regulation (eg bribery between the enforcer and someone who wants to bypass the regulation) is not a good choice.

comment by ChristianKl · 2020-11-09T17:15:56.123Z · LW(p) · GW(p)

I didn't ask anything about viability. I asked about value. It's good to first understand what's happening before then deciding about what should happen.

Bypassing regulations is very rarely progress

Why? Why isn't it progress to use technology that was previously forbidden by regulation? Using Uber instead of taxis seems to me like progress.

Replies from: Ericf
comment by Ericf · 2020-11-09T19:28:36.400Z · LW(p) · GW(p)

Dumping SO2 into the atmosphere isn't progress. Delivering cars without seat belts isn't progress. Building a factory without guard rails on the walkways isn't progress. I could go on. The examples of stifling regulation, even on a cursory glance, are few and far between.

Replies from: ChristianKl
comment by ChristianKl · 2020-11-09T19:49:45.131Z · LW(p) · GW(p)

Do you believe that increased regulatory burden isn't part of the reason for the Great Stagnation? 

Replies from: Ericf
comment by Ericf · 2020-11-10T19:41:15.154Z · LW(p) · GW(p)

I don't see how that's relevant to the original question. "Does X contribute to Y" is investigating "why care about X" "Can Z mitigate X" is investigating "how can we affect X" The question of "is it worth the costs of Z in order to reduce Y" is a higher level question. Analyzing it in a thread under the sub-question is the wrong place to have such a discussion.

Replies from: ChristianKl
comment by ChristianKl · 2020-11-10T23:59:07.769Z · LW(p) · GW(p)

It's a core assumption on which the case I made is build. 

The general thesis here is: 

  1. Regulatory burden is a main factor for the lack of progress we see in the Great Stagnation 
  2. Corruption helps actors get around regulatory burden 
Replies from: Ericf
comment by Ericf · 2020-11-11T01:26:58.483Z · LW(p) · GW(p)

I'm not commenting on either of those theses. Just pointing out that there are substantial costs to moving corruption from "verboten" to "kinda not ok"

In my estimation, the cost/benefit here is like burning your furniture for heat.

Replies from: ChristianKl
comment by ChristianKl · 2020-11-11T01:46:47.791Z · LW(p) · GW(p)

Just pointing out that there are substantial costs to moving corruption from "verboten" to "kinda not ok"

Pointing that out is pretty worthless as it's already general knowledge. 

Replies from: Ericf
comment by Ericf · 2020-11-11T02:13:34.096Z · LW(p) · GW(p)

Is it? You apparently didn't take "most regulation is not stifling" into account when you made your original post. And you gave the impression that you think the relative cost benefit is somewhere in the realm of reasonable (by even making the post), which implied that you might not be accounting for all the costs of "using corruption"

Replies from: ChristianKl
comment by ChristianKl · 2020-11-11T12:38:51.186Z · LW(p) · GW(p)

I didn't spent time talking about how the sky is blue either.

There was no intent in this post into accounting for all the costs of corruption but whether or not corruption is benefitial for a specific purpose. 

When a topic is quite charged by preconceptions it's very useful to focus on the part that's unclear to gather more understanding and not on the parts that aren't.

Replies from: Ericf
comment by Ericf · 2020-11-12T14:06:02.883Z · LW(p) · GW(p)

It's important to explicitly say that at the top of your post. That goes for any morally questionable discussion. If you don't put a disclaimer like:

"Notwithstanding the facts that 1) the first order effects of regulations are to save lives and improve the physical and mental health of the people protected by them 2) allowing corruption for some things weakens protections against corruption in other areas; let us consider ..."

As the first part of your post, reasonable people might think you either didn't know those things (because nobody knows everything, and nothing is known by everyone) or didn't believe it (because millions of people act in a way consistent with believing the opposite, and you could be one of them.)

answer by George3d6 (George) · 2020-11-08T16:37:33.299Z · LW(p) · GW(p)

Corruption can help solve problems of disproportionate blame vs credit for politicians and other functionaries.

That is to say, assume a given PS will get much more blame breaking/removing a regulation if things go south than praise if they go well. In that case, if an ambitious project needs to get rid of said regulation, they must balance the PS's cost matrix (e.g. break this regulation for us, and here's some money, plus extra if things go well).

On the other hand, the kind of corruption the US (and in part the EU) has works on a reverse model, where it's the government that gives money to businesses in order to get help passing policies, rather than the businesses that pay officials to get rid of regulations. (E.g. the way the US does healthcare, college, military equipment contracts, or space-exploration).

These two types of corruption seem completely unrelated to met. So really, I think the answer is "Yes, but, a narrower definition of corruption is needed to make the point stick"

comment by ChristianKl · 2020-11-08T18:49:33.962Z · LW(p) · GW(p)

It's unclear to me what PS stands for. It's likely not useful to use that abbreviation. 

Replies from: George3d6
comment by George3d6 · 2020-11-09T00:28:15.792Z · LW(p) · GW(p)

Meant to say public servant, no idea why I used that abreviation, sorry.

8 comments

Comments sorted by top scores.

comment by Mo Putera (Mo Nastri) · 2020-11-08T14:07:06.978Z · LW(p) · GW(p)

I'm reminded of Janus Dongye's Quora answer to "How much corruption is there in China?". It's long and informative, but his point is basically that corruption both attracts talent and gets people moving, and that rising in the CCP entails learning how to balance corruption, risk (to your reputation if things go south) and efficiency to get things done and win promotion. 

Replies from: ChristianKl
comment by Kenny · 2020-11-09T22:00:04.890Z · LW(p) · GW(p)

Would you consider Uber or Airbnb to have engaged in corruption?

I'm not sure what connections you think exist between Robert Mose's corruption and him building a lot of things. I can't think of any off-hand.

Generally, some examples would be helpful; perhaps both central or prototypical ones as well as some you'd consider to be edge cases.

Replies from: ChristianKl
comment by ChristianKl · 2020-11-09T23:09:40.660Z · LW(p) · GW(p)

I'm not sure what connections you think exist between Robert Mose's corruption and him building a lot of things.

When Robert Moses wanted things to get done, they got done. A lot of the examples of how Robert Moses used his power are to complex to explain them in a paragraph. 

That complexity is part of the way he defended his power because it was hard to understand how it was wielded. 

On example was Moses getting a politician to pass a law that meaned that Moses could take someone's house and start building a road the next day. He did this by writing the amendment in a way that neither the politician he proposed the law nor anyone in the parliament understood what the law would actually do.

If you can start building a road directly after you told someone that you want the road where their house is and don't have to wait a year for a legal process about whether it's okay to take the house of the person, it's easier to build things. 

That was in Robert Moses early days when he didn't have a lot of people directly on his payrol but it's a nice example of him doing things that allow his projects to move fast that would never go through today. Later examples are more complex because it's the web of relationships he build that prevented opposition to any project he wanted to do. Caro's book is excellent if you want to understand how power can be wielded.

Would you consider Uber or Airbnb to have engaged in corruption?

I don't know. Both started by simply ignoring the law in many jurisdictions. Later they hired political insiders to lobby. From the outside it's hard to know whether or not those political insiders use means that we should call corrupt to get laws changed. We might know more in one or two decades where someone writes a good book about what happened in their legislative efforts.

They don't seem to be powerful enough in a state like California to get the law reducing their ability to hire independent contractors from getting passed which suggests that they do have some inability to corrupt that legislative process.

Replies from: Kenny
comment by Kenny · 2020-11-11T06:08:41.869Z · LW(p) · GW(p)

Okay, that's along the lines of what I remembered from reading reviews of Caro's book – Moses's 'corruption' was, at least generally, atypical, e.g. not bribery or blatantly abusing his power. It was in a sense a much more subtle form of corruption, in large part because it wasn't obviously corruption at all.

My problem with this line of thinking is that indicts basically all political actors:

That was in Robert Moses early days when he didn't have a lot of people directly on his payrol but it's a nice example of him doing things that allow his projects to move fast that would never go through today. Later examples are more complex because it's the web of relationships he build that prevented opposition to any project he wanted to do.

You just described basically every politician. While it's certainly common to complain 'all politicians are corrupt', it's a kind of 'fallacy of grey' for eliding any degree or detail of how corrupt various acts are.

Would you consider Uber or Airbnb to have engaged in corruption?

I don't know. Both started by simply ignoring the law in many jurisdictions. Later they hired political insiders to lobby. From the outside it's hard to know whether or not those political insiders use means that we should call corrupt to get laws changed.

Right – that's what I was getting at because I thought your conception of 'corruption' was too vague and too expansive.

Is "simply ignoring the law" corrupt?

Is hiring lobbyists corrupt?

Is lobbying itself corrupt? (I'd imagine you'd agree it's not, apart from 'stereotypical lobbying', and maybe even not then.)

They don't seem to be powerful enough in a state like California to get the law reducing their ability to hire independent contractors from getting passed which suggests that they do have some inability to corrupt that legislative process.

Is the ability to influence a bill being passed, and exercising that ability, corruption? Is it only corrupt if you're a corporation or its agent?

I'm not sure what you mean by 'corrupt' in "corrupt that legislative process". It seems like any effect they caused on the bill passing or not would be corruption. That doesn't seem right.

Replies from: ChristianKl
comment by ChristianKl · 2020-11-11T14:46:08.328Z · LW(p) · GW(p)

Moses's 'corruption' was, at least generally, atypical, e.g. not bribery or blatantly abusing his power. 

He also did engaged in less atypical actions. He hired people to dig up dirt on his opponents. Once he had the Triborough Bridge Authority he used that to pay people off. It was however more paying people for to be loyal to him in general instead of for one particular project.

I thought your conception of 'corruption' was too vague and too expansive.

There's a reason this post is a question and not an essay.  I point to something unclear that I don't understand well.

Is the ability to influence a bill being passed, and exercising that ability, corruption? Is it only corrupt if you're a corporation or its agent?

The standard working definition of corruption I use is that it's about trading things of different domains. If Alice votes with Bob on bill X in return for Bob voting with Alice on bill Y that's not corruption but normal negotiation of interests in politics.

If Alice votes with Bob on bill X in return for Bob giving her personally cash, that's corruption. 

A government drafts a new law and publishes a request for feedback. A lobbyist uses his expertise to answer the request for feedback and explains a way how the law can be improved. If the government then finds that answer convincing on it's merits and incorporates the feedback that's not corrupt. 

When talking to a ex-lobbyist they explained to me that this was a lot of what they did in their work and that's white-hat lobbying that's important for having a government that doesn't accidently pass laws that kill part of it's industry because the government doesn't understand what the laws they propose do.

If Elon Musk goes to the Texan government and says "I would like to launch rockets from Texas, at the moment the laws about noise pollution forbid this. Can you change the laws in a way that allows me to lunch rockets from Texas? I would employ Texan citizens to work for me" that's a reasonable request. It would be reasonable if the politician on the other side asks "How many Texan citizens would you employ? Can you promise to employ at least X number of our citizens in SpaceX?"

You want an entrepreneur like SpaceX to be able to start such an exchange even without a request for feedback. 

If the politician however says "My son would like an internship at SpaceX, can you see that he gets that?" that would be corrupt as it's trading things of different domains. 

There's some debate about whether or not campaign donations should count as being in the political domain and thus fair game or not. While they are less corrupt then direct personal benefits, I think it still makes sense to see them as corruption. The fact that in the above example Elon Musk would have likely given some campaign donation before getting the meeting to discuss his proposal is a form of corruption and this paying for access is widespread in Washington. 

With Uber there are allegations that Uber used his own systems to track politicians. If the information from tracking politicians was used to influence the political process that's corrupt as it's using things of different domains.  

Uber/AirBnB getting what they want in legislation would not be evidence of corruption but them not getting what they want is evidence that the lack power to corrupt the system in the way they want to.

Replies from: Kenny
comment by Kenny · 2020-11-13T03:03:44.865Z · LW(p) · GW(p)

There's a reason this post is a question and not an essay. I point to something unclear that I don't understand well.

Sorry if it wasn't clear but I'm also trying to make sense of your question and understanding what you mean by 'corruption' was my most significant obstacle.

The standard working definition of corruption I use is that it's about trading things of different domains.

My problem with this is that it still seems too nebulous. I'm not sure how to answer the question without having a sharper distinction for what is and isn't 'corruption' so as to at least estimate its total costs and benefits.

You mentioned "white-hat lobbying" and that's exactly what I was thinking of when I asked whether you consider lobbying itself to be a form of corruption. From the evidence and info I've gleaned from the lobbyist I know, there's quite a bit of the 'mechanics' involved, e.g. in 'accessing' politicians, that seems at least a little corrupt.

And I've read a reasonable defense of corruption in general arguing that, in effect, corruption provided something liked liquidity to political markets and that made negotiation among coalitions generally easier to conduct (which is itself arguably good). I'm very sympathetic to that.

I'm sorry I haven't been able to provide a clear answer or even clear thinking about your question. I think it's an interesting question tho!

comment by Kenny · 2020-11-09T21:52:49.861Z · LW(p) · GW(p)

This is a great question!