Laws Don't Always Work as They Should in a Complex World

[Fotheringay-Phipps]

From the Washington Post: The law was supposed to reduce discrimination. But it made hiring more racially biased.

[…] one of the hottest ideas among lawmakers right now is to ban employers from running credit checks on job applicants. Since 2007, eleven states, as well as Chicago and New York City, have passed such laws.

Supporters of these restrictions often frame the issue as a civil rights problem. In particular, they say, credit checks impede employment among minorities, who disproportionately have low credit scores.

But a new study from Robert Clifford, an economist at the Boston Fed, and Daniel Shoag, an assistant professor at Harvard’s Kennedy School, finds that when employers are prohibited from looking into people’s financial history, something perverse happens: African-Americans become more likely to be unemployed compared to others.

And as for drug tests …

A powerful study published last year in the Review of Economics and Statistics shows something of the opposite happening: When employers began to require drug tests for job applicants, they started hiring more African-Americans.

“The likely explanation for these findings is that prior to drug testing, employers overestimated African-Americans’ drug use relative to whites,” the study’s author explained in an op-ed. Drug tests allowed black job applicants to disprove the perception that they were addicts.

The general lesson to be learned, IMO, is that the world in general and human behavior in particular are very complex, and it’s often hard to predict what the overall impact will be of laws that seek to regulate them. While it’s often tempting to pass laws that ostensibly tackle problems head on, and these can make lawmakers feel they’re solving problems and appeasing popular sentiment at the same time, the reality might be very different.

As a practical matter, the relevance is, IMO, that there should always be a bias in favor of not passing any proposed new law over passing one, especially if the area it seeks to regulate is a clearly complex one.

Don’t do something! Just stand there!

[HurricaneDitka]

So are you a libertarian now?

[Senegoid]

What’s missing is a relatively efficient legislative mechanism for fixing or fine-tuning existing laws.

Congress passes big complex laws (for purposes of this discussion, let’s assume that we actually have a do-something Congress), then moves on to other big complex issues. Fine-tuning existing laws is typically seen as low-priority penny-ante stuff that nobody cares to bother with.

So you get all these laws with their unintended consequences, and the necessary fine-tuning just doesn’t happen much. We need a mechanism within our legislative/jurisprudence system to facilitate that.

Say, for example, a standing committee in Congress whose whole charter is to follow up on existing laws, observe how they are working, collect data on unintended consequences, and propose new legislation to amend existing laws to fix the problems, which the full House and Senate should then be able to pass without too much fuss or controversy (hopefully). There currently just isn’t much incentive for anyone to do that.

[PatrickLondon]

One thing, from a foreign perspective: it seems to me that your politicians tend to micro-manage through writing into law what we would treat as administrative interpretation of law. To take the OP, we have anti-discrimination laws, but as to whether running credit checks or drug tests on job applicants would amount to indirect discrimination wouldn’t be the subject of specific statute law or regulations. If it needed to be challenged as such, it would be tested in the employment tribunals in the light of the evidence.

But I do agree that there is a law of unintended consequences. Laws against prostitution are a classic example - every time there’s been a grand review and reformed legislation brought in to deal with some perceived nuisance, a new one is created somewhere else along the line. There used to be a character on a children’s radio programme years ago who was always getting annoyed and shouting “It Ought Not To Be Allowed!”; that, and “Something Must Be Done”, are the most ominous words in political discourse, it seems to me.

[adaher]

Another similar problem is requiring SS#s for job applicants. This law was passed in 1986 and was supposed to make it harder for illegal immigrants to get jobs. Instead of making it harder for illegal immigrants to get jobs, it basically created the identity theft industry.

What this whole thing shows is that politicians aren’t all that bright, and whether or not to admit they were wrong depends entirely on the political circumstances, not whether a law needs “fine tuning”, as Senegoid put it. The system works just fine for fine tuning laws, if we assume a legislature that acts in good faith for the benefit of the citizenry. Instead, these laws were all about politics. Whether or not they work is irrelevant. The politicians sent a message. Their work is done.

[UDS]

Societies are complex, and pretty well all laws have unintended effects (though, often, they’re not problematic and we’re happy to live with them).

I disagree that the result would be an instinctive aversion to legislating. I take the point that that would have avoided laws banning discrimination on the basis of credit scores, but the same instinct would tend to avoid laws banning discrimiantion on the basis of race (which we all agree, don’t we, are good laws that we are right to have?).

I think we need to do two things. First, think about laws before we enact them, and test them against the evidence. Accept that the outcome of the law that you are hoping for will not be the only outcome, and put a bit of thought into identifying and assessing other likely outcomes. Laws should be based on well-thought out and well-tested policy, in other words.

Secondly, accept that laws do have unintended outcomes and this is unavoidable. So, as Senegoid suggests, part of any robust legislative process has to be review of the operation of the law once implemented, followed by repeal or modification as appropriate.

And if you need to repeal or modify the law, don’t regard that as necessarily a failure. You framed your policy, you implemented it, you tested it and you have learned something. If the law had disastrous outcomes that really ought to have been foreseen yes, that’s a failure. But if the unintended outcomes were the result of changing circumstances, or factors that couldn’t reasonably have been identified or assessed when the policy was first made, then you have learned something useful.

But the real problem here is, well, democracy. One of the main reasons why laws have consequences which are not those desired is that the laws are popular. Get-tough-on-crime laws, for example, are popular and get passed for that reason even though we know that, mostly, they are not very effective to reduce crime rates and the principal outcome is to increase taxes to pay for more prison accommodation. But the problem isn’t that nobody foresees this; it’s widely recognised. So while these consequences are perhaps unintended, they’re not unexpected. And that problem is not down to poor policy preparation or a failure to review the legislation once implemented.

[adaher]

In other words, legislatures are motivated by politics. Everything else is a very distant second.

[UDS]

@adaher

In other words, legislatures are motivated by politics. Everything else is a very distant second.

Yes, but this isn’t a complete disaster since, generally, framing good policy and delivering good outcomes are politically advantageous things to do. The problem is only an acute one where what is popular and what is effective are opposed.

[adaher]

You’re onto something there. The problem is that the politicians have designed the system so that accountability is hard to pinpoint. In the most general sense, politicians who already have jobs want a satisfied public, so avoid either big changes that upset too many apple carts or damage the economy. But on these types of wedge issues, like telling businesses they can’t do credit checks, that was done to make a statement, not to get any particular result. I’m seeing no movement to fine tune or touch the policy in any way, and I don’t think we will see it. The problem I mentioned with SS#s being required for employment has been well understood for 25 years and the only thing the government has done, on the downlow, is make sure that those who have their identities stolen aren’t informed so that the government can collect the extra tax money.

[bldysabba]

@UDS

Yes, but this isn’t a complete disaster since, generally, framing good policy and delivering good outcomes are politically advantageous things to do.

You would think so, but the evidence for it is remarkably scant.

[davida03801]

There are a couple things that legislatures all over the world cannot repeal. The Law of Unintended Consequences is one of them.

[wolfpup]

@Fotheringay-Phipps

As a practical matter, the relevance is, IMO, that there should always be a bias in favor of not passing any proposed new law over passing one, especially if the area it seeks to regulate is a clearly complex one.

Not having a law is every bit as much of a factor in governing behavior as having one. Is it somehow axiomatic that the absence of a law always leads to the better outcome?

This is exactly like the argument that says that government is capable of being corrupt and stupid, and therefore the solution is to have minimal government. Actually, no, the solution is to have honest and competent government. Because a functional government is the only thing that stands between you and all the evils and injustices that naturally arise among the self-serving powers of society.

Your argument is nothing more than a restatement of the libertarian’s prime article of faith.

[Flyer]

@Senegoid

Congress passes big complex laws (for purposes of this discussion, let’s assume that we actually have a do-something Congress), then moves on to other big complex issues. Fine-tuning existing laws is typically seen as low-priority penny-ante stuff that nobody cares to bother with.

The attitude shown in the first half of this paragraph is the reason for the problem that you lament in the second half.

Far too many people assume that the sole purpose of Congress is to pass laws (“a do-something Congress”). We have too many laws already. The purpose of Congress is to govern. Contrary to popular opinion, that does NOT simply mean passing laws–it means to act in accordance with the Constitution, and to pass laws only when there is a genuine need for a new law.

[Really_Not_All_That_Bright]

I don’t understand what the problem is here. Lawmakers identified a problem. They crafted a solution and enacted it. It hasn’t worked, or at least one study says it hasn’t worked. Assuming that is the case, they can repeal the law.

This is how democracy is supposed to work.

[Grumman]

@wolfpup

Not having a law is every bit as much of a factor in governing behavior as having one. Is it somehow axiomatic that the absence of a law always leads to the better outcome?

Always by default. If it’s a coin toss whether a new law is better or worse, it should never be passed, because all things being equal a new law is always worse than not having a new law. It is worse because it is a new infringement upon people’s ability to act as they see fit, and such an infringement needs to be justified by something better than “eh, it might be worth it”; and because it is a change to the rules you are demanding the people know, understand and abide by, and every time you do that you make it harder for people to abide by the law even if they want to.

[Blaster_Master]

@Grumman

Always by default. If it’s a coin toss whether a new law is better or worse, it should never be passed, because all things being equal a new law is always worse than not having a new law. It is worse because it is a new infringement upon people’s ability to act as they see fit, and such an infringement needs to be justified by something better than “eh, it might be worth it”; and because it is a change to the rules you are demanding the people know, understand and abide by, and every time you do that you make it harder for people to abide by the law even if they want to.

Not just a coin toss, but it should have a clearly positive result including the costs, adminstrative burdens, reductions in liberty, and the risk of unintended consequences. So even if it’s just mildly better than a coin toss as whether or not it’s a benefit, it probably isn’t. And these sorts of things aren’t easy to judge, especially the risk of unintended consequences.

Using the OP’s example, I get that there’s a correlation of minorities having low credit scores, but rather than prevent potential employers from running a credit check, which admittedly is silly for some jobs but reasonable for others and which reduces their available information in making an informed hiring decision, why can’t we instead find a way to fix the issues with credit scores?

Hell, I have a really good credit score, and I’m still not very clear exactly what behavior does and doesn’t affect it, and how much it does. I know people that make good money and are reliable, but have made a couple minor mistakes (e.g., missed an auto-payment because they changed banks) or took too long to establish credit, and it kills them. Even ignoring it’s affect on a job, a credit score does a lot to determine where we can live, what car we drive, what school we may attend, and other essentials or non-essentials we can afford.

In short, it seems like a lot of these laws that have downsides are often well intended, but poorly thought out. And, really, more of these sorts of laws should be easier to repeal, or have sunset clauses, so they automatically go away unless they are working, at which point there’s evidence to support that.

[davidm]

Welcome to the real world. Anything of any complexity is likely to have unintended consequences, and some of those consequences may be negative. This is not unique to governance.

In spite of this, most of us still get out of bed in the morning and keep trying.

[YogSothoth]

@Fotheringay-Phipps

Don’t do something! Just stand there!

This assumes there’s no problems in the first place, which is patently false.

Where laws screw up with unintended consequences, lawmakers should fine tune the law to adhere to the original goal and weed out negative consequences. Only a person originally opposed to fixing the problems at all would claim that the issue is actually lawmakers attempting to fix something, and shame them into stopping, rather than tackle the problem with a solution in the first place. Given your posting history, I’ve no doubt which side of the fence you stand on

The goal is diversity, equality, and fairness. If law X doesn’t do it, change it into law Z, repeat as many times as necessary

[adaher]

@Really_Not_All_That_Bright

I don’t understand what the problem is here. Lawmakers identified a problem. They crafted a solution and enacted it. It hasn’t worked, or at least one study says it hasn’t worked. Assuming that is the case, they can repeal the law.

This is how democracy is supposed to work.

This is where democracy breaks down, actually. Some sensible people will call for its repeal(REpublicans), and emotional people will say that Republicans want to make it easier for employers to discriminate(Democrats).

[adaher]

@wolfpup

Not having a law is every bit as much of a factor in governing behavior as having one. Is it somehow axiomatic that the absence of a law always leads to the better outcome?

This is exactly like the argument that says that government is capable of being corrupt and stupid, and therefore the solution is to have minimal government. Actually, no, the solution is to have honest and competent government. Because a functional government is the only thing that stands between you and all the evils and injustices that naturally arise among the self-serving powers of society.

Your argument is nothing more than a restatement of the libertarian’s prime article of faith.

Government is a necessary evil, because government involves the use of force, which is also not only a necessary evil, but should be utilized as little as possible. The argument for minimal government isn’t some libertarian article of faith. It’s a very well reasoned argument. Liberals tend to elide this argument by seeing the use of potentially deadly force as a positive good as long as it’s used in the service of a cause liberals support, like preventing people from selling loose cigarettes on the streets of NYC. Then they decry the use of force and call for these little laws to not be enforced. Then the whole concept of rule of law gets undermined. Just don’t pass laws you aren’t willing to kill for in the first place.