Unintended consequences hamper the efforts of well-intentioned policy-makers everywhere, whether they choose to recognise them or not. They occur due to issues inherent to the implementation of said policies. The main issue, widely cited among classical economists and roundly ignored among other schools, is the Friedrick Hayek's signature "knowledge problem."
Hayek popularized the notion of a knowledge problem burdening central planners in his seminal essay, "The Use of Knowledge in Society." In it, he discussed how central planners, who oversee economic activity from a high perch, necessarily have less knowledge about particular local events and situations than economic actors at ground level. Therefore, the means by which central planners can achieve benevolent outcomes is necessarily limited.
This problem is worsened by Hayek's observation that most knowledge is situational and descriptive, not statistical or strictly based on easily-communicable facts. So while a naturalist would have an easy time listing the height, weight and attributes of a particular bird species, it would take a lifetime for him to understand a particular bird's personality and environment so well that he could predict it's actions. Extrapolate the problem of understanding one actor, the bird, to an entire economy of actors, and you'll see why economic planning is such a Sisyphean task. This also helps explain why economic forecasting is notoriously unreliable.
The problem for economic thinkers and planners is not strictly one of methodology. An economist who conducts his forecasts with a perfect methodology would still be handicapped by the ignorance that Hayek described. Whatever economic facts they were able to obtain would be finite, while the knowable information would be infinite.
This problem extends, naturally, to regulators. Economic policy is based finite knowledge but comes with infinite implications. This observation does not, however, imply that regulation(or for that matter economics) is a worthless endeavor. Instead, it merely implies that a regulator has limited tools to achieve benevolent outcomes. Therefore, the number of problems which they should attempt to correct should also be limited. Among them are market failures, such as prisoner's dilemma problems, where market actors tend toward poor outcomes in spite of superior knowledge. The consequences of an intervention in such cases, in spite of their unforeseeable nature, will have a fighting chance of being positive,