"Structuring successful policy interventions is a surprisingly difficult task"

Well, duh (Economist):

The authors focus on an experiment run by the British government in the early 2000s: an RCT that offered incentives to disadvantaged people to stay in work. A group of 3,500 single mothers, who were either out of work or working part-time, were split into two groups. The control group was given nothing; the treatment group was given coaching and financial incentives to work. Individuals working more than 30 hours per week, for example, could receive a tax-free payment of £400 ($666). All forms of treatment lasted, at best, just under 3 years.

Five years after the experiment started, the authors checked up on their subjects. The treatment group reported significantly lower levels of well-being, even though those individuals ended up with higher earnings than the control group. The treatment group were less happy with their lives and worried more about budgeting and debts. Helping people, in other words, seemed to hurt them.

Why? The results may be to do with the quality of the training on offer. Even if people end up earning more, the experience of being bossed around by a uncaring counsellor may make people unhappier.

Another plausible explanation is, in layman’s terms, related to unrealistic expectations. People in the treatment group enjoyed the advantage from taking the government subsidy. The extra money probably allowed them to buy better things and worry less about making ends meet. But the subsidy came to an end. When it did, the control group found living without the subsidy harder to bear. They had to live with the curse of raised aspirations.