Maybe the average amount of the fine needs to be more than the value of the societal impact.
The purpose of a fine can be seen in at least two ways:
1. Provide recompense to society for the production of a negative externality.
2. Create a disincentive against the production of a negative externality.
If fines could be levied perfectly, automatically and without administration cost, then perhaps you can see the fine as #1, and set the fine at the societal cost.
But, because it costs money to enforce regulations, there's usually some sampling done: if you park illegally, you might not be caught. Because the chance of getting caught is not 100%, and because those who are not caught do not contribute their share of enforcement costs, the fines for a single infraction must be a multiple of the actual harm caused. (Assuming that can even be measured.)
Let's say the harm caused is $20 per infraction, and this is normally enough to disincentive a poor person from parking illegally. If they make a mistake and are subsequently caught, is it better for them to pay $100 (for the harm caused, plus paying for those who weren't caught), even though this may be equivalent to their family's monthly food budget?
As long as $20 fine is enough to:
- provide a strong disincentive, and
- provide recompense
... then that should be enough, no?
The rich person, for whom a $20 fine is not enough of a disincentive, should pay more ($200?). And the money raised can be used to cover negative externalities for those who weren't caught, and overall enforcement costs.
Taken seriously, this would be a very complex system and it would fail to realize its claimed potential.
It only sounds good in low resolution.
Depending on a bunch of hard questions, devils in the details, you’ll end up with dilemmae where neither option is good:
Is parking income net worth, or taxable income? Previous year or trailing? Can you deduct your charity donations before deciding what you’ll pay for parking? What about a company car; is it the company’s income or the driver’s?
We are techies, we can always build a more complicated system; it is our nature:)
A more simple system off the top of my head: you have fine modifier based on your previous year taxes. Your accountant now tells you you have a fine multiplier of 6. When the ticket comes in, it is always addressable to a given entity. Can it be gamed? Of course, just use a car registered to your broke ass cousin. Progressive fines could work just like progressive tax.
The purpose of a fine can be seen in at least two ways:
1. Provide recompense to society for the production of a negative externality.
2. Create a disincentive against the production of a negative externality.
If fines could be levied perfectly, automatically and without administration cost, then perhaps you can see the fine as #1, and set the fine at the societal cost.
But, because it costs money to enforce regulations, there's usually some sampling done: if you park illegally, you might not be caught. Because the chance of getting caught is not 100%, and because those who are not caught do not contribute their share of enforcement costs, the fines for a single infraction must be a multiple of the actual harm caused. (Assuming that can even be measured.)
Let's say the harm caused is $20 per infraction, and this is normally enough to disincentive a poor person from parking illegally. If they make a mistake and are subsequently caught, is it better for them to pay $100 (for the harm caused, plus paying for those who weren't caught), even though this may be equivalent to their family's monthly food budget?
As long as $20 fine is enough to:
- provide a strong disincentive, and
- provide recompense
... then that should be enough, no?
The rich person, for whom a $20 fine is not enough of a disincentive, should pay more ($200?). And the money raised can be used to cover negative externalities for those who weren't caught, and overall enforcement costs.