3 ms·
> laws should be agnostic to whether the technology is emerging or traditional In a world where everyone can anticipate the consequences of everything, this wo
by speedplane 7y ago
> laws should be agnostic to whether the technology is emerging or traditional
In a world where everyone can anticipate the consequences of everything, this would be correct. However, such a world does not exist.
Traditional technology has the benefit of having known social cost/benefit consequences. Emerging technology has no track record, and there is not enough data to perform a proper cost benefit analysis.
A relevant example: Cities have a lot of experience regulating taxis and analyzing how they affect traffic, transportation, public safety, and cost of living. In contrast, the consequences of adding scooters to the streets are unclear. They likely improve transportation options, but they may also slow down vehicular traffic, create public safety hazards, and siphon money away from public transportation systems. A comprehensive cost-benefit analysis may eventually show scooters are a net positive or a net negative. However, because it's an emerging technology, there is not enough data to do such an analysis.
When there is less confidence and data, it's entirely reasonable to take that uncertainty into account when drafting policy. Some localities may choose to be faster adopters, a higher risk tolerance for risk in exchange for faster innovation. Others may have a lower tolerance for risk, preferring to value stability over quick improvements.
Reasonable people can disagree on the appropriate risk tolerance, but regardless, risk of the unknown is still a legitimate factor to weigh when crafting policy.
- CryptoPunk 7y ago>>In a world where everyone can anticipate the consequences of everything, this would be correct. However, such a world does not exist. That's the wrong way to approach the issue of what laws to craft. The action itself should be judged on its own merit, in its cost to society, independent of its connection to other factors that may have negative or positive impacts on society. The cost to society of an object being place in a pedestrian path way should be measured, regardless of whether that object is a scooter that improves transportation options or a rock that doesn't. If those imposing negative externalities are made to stop, or made to pay for them, then only those activities that have a net positive impact on society will be economically sustainable. With the scooter example, if the benefits scooters provide in increasing mobility outweigh the charges levied on scooter rental companies for the costs of increased sidewalk congestion, then the business model will be viable. The cost-benefit analysis should be done by entrepeneurs, not by one set of regulators making all decisions. The regulator's job should be to calculate the negative externalities each action imposes and ensure entrepeneurs pay for them. That in turn will align the financial success of entrepreneurs with the social good. Each party confining itself to its own role is a more scalable way to organize society than the state attempting to model every factor as part of a large-scale cost-benefit analysis and effectively make business decisions that should be made by numerous private investors with their own capital on the line for their respective business. Central economic planning is not an effective way to organize and run society. >>Cities have a lot of experience regulating taxis and analyzing how they affect traffic, transportation, public safety, and cost of living. If cities knew what they were doing, there wouldn't be so many poorly run regulated taxi services.
- speedplane 7y ago> The action itself should be judged on its own merit, in its cost to society, independent of its connection to other factors that may have negative or positive impacts on society. I think you're missing my point. I agree that many policies should largely be measured with an objective cost benefit analysis. However, for emerging technologies, it can be very difficult to accurately perform such a cost benefit analysis. There is less data, and confidence levels are lower. It's much easier to evaluate the cost/benefit of taxis than scooters simply because there is far more historical data.
- CryptoPunk 7y agoMy point is that it should not be the job of regulators to do cost-benefit analyses on product categories. It should be to simply measure the cost of a negative externality, in a technology/product agnostic way, and make sure entrepeneurs pay that cost. The cost-benefit analysis is then done by private parties, in order to avoid losses and maximize profits.
- speedplane 7y ago> The cost-benefit analysis is then done by private parties, in order to avoid losses and maximize profits. This is such a theoretical statements that ignores so many realities. Uber wants to enter a new city. The city has an established taxi industry and policies in place to manage pollution, congestion, housing, and new development. What "private parties" are supposed to do the cost-benefit analysis here? Obviously it's in Uber's interest to move in, but what if that increases pollution or the livability of the city to the existing residents? There are no private parties that are in a position to do a credible and representative analysis of the costs and benefits.