4 ms·

Yeah. we hire ex regulator employees cos we are trying to comply with regs, not beat them. regulations and regulatory feedback can be hard to understand, use a

by retube 8y ago

Yeah. we hire ex regulator employees cos we are trying to comply with regs, not beat them. regulations and regulatory feedback can be hard to understand, use arcane language, and may be apparently inconsistent. having people who can interpret and advise is very useful.

justaguyhere 8y ago
having people who can interpret and advise The solution here is not to hire those who wrote those regulations at massive salaries. The solution is to write the regulations that is easy to interpret by those for whom it was written in the first place! This is like me writing un-maintainable, insane code that only I and I alone can understand and nobody else, just so I can keep my job! This would be the fastest way for me to get fired. But somehow this is the norm in law and regulations?
- hndamien 8y ago
  English is apparently a horrible language for writing law. Especially bad for reading it.
  - justaguyhere 8y ago
    I wonder if this is the case in other English speaking countries like England, Australia etc? This sounds like a flimsy excuse...
    - hndamien 8y ago
      I was half joking, like people talking about programming languages. Legal docs just aren't very user-friendly have pretty large attack surfaces due to language.
- sct202 8y ago
  I think the challenge with regulation is that it is infrequently revised, so imagine you have to write code that is works and also cannot be edited for 10 years. It's a long time, so if you happen to over specify your regulations, you might end-up creating a giant loophole that will be open for years. If you under-specify it, you end up with a lot of uncertainty over what is or isn't included. Having been on the receiving end of a lot of safety and environmental regulations, I would have appreciated more specific guidelines, especially with the number of meetings I had to go to where we would internally try to justify if our decisions were going to violate something or not.
- pc86 8y ago
  > But somehow this is the norm in law and regulations? If you want to willfully misinterpret the comment to which you're replying, sure. Think of it more like this. You write shit code not because you want to, but because the other code it needs to run with is similarly shitty and perhaps you have laws requiring it be a certain level of shit. You write this shit code for 60% of the salary of your former classmates, with albeit marginally better benefits than you could buy for that extra 40% on the public market. This shit code is used almost exclusively by one sector of the economy. The companies in that sector of the economy have two choices when hiring people: 1. Hire someone like you who is intimately familiar with all the technical and business debt of the shit code you spent the last two decades writing. Because it's so complicated, and because you've made comparatively little money during your career, and you have a pension now, you're not really interested in doing this for less than $250-300k a year. 2. Hire someone else who needs to learn all this on the job. But everyone else is hiring people like you so "market rate" for this gig is at least $200k for the guys with no experience relying on a connection or two to even get the interview. It's easy to say "well just do $x" while ignoring that $x is not realistically possible.
- ksdale 8y ago
  I agree that there is regulation that is needlessly complicated, but there are also a lot of regulations that are only perceived that way because they need to be complex to cover a bunch of important edge cases. I can't count the number of times in law school that someone would raise a point about something being written ridiculously and the professor would walk them through an alternate way of writing it, and by the end, you often saw that what appeared convoluted was actually a fairly elegant solution. I definitely came to law school with a very libertarian perspective and left with a grudging respect that most laws are actually written fairly optimally. Whether they should exist is a different question, but given that they exist, it is generally unlikely that there is a simpler way to write the law. The fact that something is complex doesn't mean it is needlessly complicated. All that said, I think there should be some sort of "cooling off period" where people can't work in a certain private industry for a number of years after working on regulation for that industry.
- tzs 8y ago
  In coding terms, writing regulations is like writing code for something that is going to be attacked by determined adversaries with an abundance of resources to use in the attack. For example, consider a simple rule that says a dividend from stock is ordinary income, and money from a sale of stock is capital gains income. It could be simpler--just count everything as ordinary income instead of shoving some off into a separate "capital gains" category that is taxed lower--but that was Congress' call, not the regulator's call. Congress has chosen to make the distinction, and so the regulator has to live with it. A long time ago, that simple rule or something like it was what was in the regulations. Then some well funded party got clever. They wanted to give their shareholders a dividend. Let's say that dividend came out per share to 1% of the value of the share. Instead of just doing that, they had the bright idea of doing a fractional stock split that takes each 1 share to 1.0102051 shares, with the price per share dropping to .989898 times the old price. That would be immediately followed by a mandatory stock buyback of 1.01021% of the shares, which would (if I did the math right when picking the split and buyback numbers) raise the share price back to the original price, and leave each shareholder with the same number of shares they had before plus cash from the buyback equal to 1% of the original share price. But now that money came from a sale of stock, not a dividend! It was now capital gains according the party's lawyers and accountants. OK, this is easy for the regulators to fix. Just add to the rule that stock buybacks are ordinary income. But then you'll run into cases where a stock buyback is actually motivated by some legitimate (from a public policy point of view) reason instead of simply trying to get capital gains status. By the time you close off all the tricks, while still giving capital gains status to buybacks that are not trying to game the system, what started out as a simple rule of one seemingly clear sentence will be a few pages delving into considerations of how wide the buyback is and whether it causes substantial changes in the way ownership is distributed among the shareholders in order to classify buybacks as legitimate or attempts at gaming the system. Now repeat this for all the other simple regulation, and pretty soon your nice simple, clear tax regulations are thousands and thousands of pages. BTW, this is why most flat tax proposals are silly. They tend to promise great tax simplification by going to a flat rate (which they do not actually propose--they almost always actually keep a progressive rate with two brackets), but the non-flat rate structure in the current US tax code has almost nothing to do with why it is not simple. The non-flat rate structure takes up something like a paragraph on the first or second page of the IRS regulations. Just going flat shortens that paragraph. Once you get past all that, it is just one or more trivial table look ups. The complexity in the tax code comes from answering the questions "how much of my income is taxable?", "how much of the taxable income is capital gains?", "what tax credits can I claim?", and so on. None of that cares whether the rate structure is flat, highly progressive, or anything else.

by retube 8y ago

justaguyhere 8y ago

having people who can interpret and advise The solution here is not to hire those who wrote those regulations at massive salaries. The solution is to write the regulations that is easy to interpret by those for whom it was written in the first place! This is like me writing un-maintainable, insane code that only I and I alone can understand and nobody else, just so I can keep my job! This would be the fastest way for me to get fired. But somehow this is the norm in law and regulations?

hndamien 8y ago
English is apparently a horrible language for writing law. Especially bad for reading it.
- justaguyhere 8y ago
  I wonder if this is the case in other English speaking countries like England, Australia etc? This sounds like a flimsy excuse...
  - hndamien 8y ago
    I was half joking, like people talking about programming languages. Legal docs just aren't very user-friendly have pretty large attack surfaces due to language.
sct202 8y ago
I think the challenge with regulation is that it is infrequently revised, so imagine you have to write code that is works and also cannot be edited for 10 years. It's a long time, so if you happen to over specify your regulations, you might end-up creating a giant loophole that will be open for years. If you under-specify it, you end up with a lot of uncertainty over what is or isn't included. Having been on the receiving end of a lot of safety and environmental regulations, I would have appreciated more specific guidelines, especially with the number of meetings I had to go to where we would internally try to justify if our decisions were going to violate something or not.
pc86 8y ago
> But somehow this is the norm in law and regulations? If you want to willfully misinterpret the comment to which you're replying, sure. Think of it more like this. You write shit code not because you want to, but because the other code it needs to run with is similarly shitty and perhaps you have laws requiring it be a certain level of shit. You write this shit code for 60% of the salary of your former classmates, with albeit marginally better benefits than you could buy for that extra 40% on the public market. This shit code is used almost exclusively by one sector of the economy. The companies in that sector of the economy have two choices when hiring people: 1. Hire someone like you who is intimately familiar with all the technical and business debt of the shit code you spent the last two decades writing. Because it's so complicated, and because you've made comparatively little money during your career, and you have a pension now, you're not really interested in doing this for less than $250-300k a year. 2. Hire someone else who needs to learn all this on the job. But everyone else is hiring people like you so "market rate" for this gig is at least $200k for the guys with no experience relying on a connection or two to even get the interview. It's easy to say "well just do $x" while ignoring that $x is not realistically possible.
ksdale 8y ago
I agree that there is regulation that is needlessly complicated, but there are also a lot of regulations that are only perceived that way because they need to be complex to cover a bunch of important edge cases. I can't count the number of times in law school that someone would raise a point about something being written ridiculously and the professor would walk them through an alternate way of writing it, and by the end, you often saw that what appeared convoluted was actually a fairly elegant solution. I definitely came to law school with a very libertarian perspective and left with a grudging respect that most laws are actually written fairly optimally. Whether they should exist is a different question, but given that they exist, it is generally unlikely that there is a simpler way to write the law. The fact that something is complex doesn't mean it is needlessly complicated. All that said, I think there should be some sort of "cooling off period" where people can't work in a certain private industry for a number of years after working on regulation for that industry.
tzs 8y ago
In coding terms, writing regulations is like writing code for something that is going to be attacked by determined adversaries with an abundance of resources to use in the attack. For example, consider a simple rule that says a dividend from stock is ordinary income, and money from a sale of stock is capital gains income. It could be simpler--just count everything as ordinary income instead of shoving some off into a separate "capital gains" category that is taxed lower--but that was Congress' call, not the regulator's call. Congress has chosen to make the distinction, and so the regulator has to live with it. A long time ago, that simple rule or something like it was what was in the regulations. Then some well funded party got clever. They wanted to give their shareholders a dividend. Let's say that dividend came out per share to 1% of the value of the share. Instead of just doing that, they had the bright idea of doing a fractional stock split that takes each 1 share to 1.0102051 shares, with the price per share dropping to .989898 times the old price. That would be immediately followed by a mandatory stock buyback of 1.01021% of the shares, which would (if I did the math right when picking the split and buyback numbers) raise the share price back to the original price, and leave each shareholder with the same number of shares they had before plus cash from the buyback equal to 1% of the original share price. But now that money came from a sale of stock, not a dividend! It was now capital gains according the party's lawyers and accountants. OK, this is easy for the regulators to fix. Just add to the rule that stock buybacks are ordinary income. But then you'll run into cases where a stock buyback is actually motivated by some legitimate (from a public policy point of view) reason instead of simply trying to get capital gains status. By the time you close off all the tricks, while still giving capital gains status to buybacks that are not trying to game the system, what started out as a simple rule of one seemingly clear sentence will be a few pages delving into considerations of how wide the buyback is and whether it causes substantial changes in the way ownership is distributed among the shareholders in order to classify buybacks as legitimate or attempts at gaming the system. Now repeat this for all the other simple regulation, and pretty soon your nice simple, clear tax regulations are thousands and thousands of pages. BTW, this is why most flat tax proposals are silly. They tend to promise great tax simplification by going to a flat rate (which they do not actually propose--they almost always actually keep a progressive rate with two brackets), but the non-flat rate structure in the current US tax code has almost nothing to do with why it is not simple. The non-flat rate structure takes up something like a paragraph on the first or second page of the IRS regulations. Just going flat shortens that paragraph. Once you get past all that, it is just one or more trivial table look ups. The complexity in the tax code comes from answering the questions "how much of my income is taxable?", "how much of the taxable income is capital gains?", "what tax credits can I claim?", and so on. None of that cares whether the rate structure is flat, highly progressive, or anything else.