Think of all those independent consumer reports on washing machines, lawn mowers and hairdryers that test ten products to breaking point and then recommend which is the best to buy? What would it be like if we had a similar report on programs and policies for children?

Or, to put it another way: what if we were to think of children's services as an investment portfolio? There is plenty of help available for those investing in stocks and shares – much of it is based on sound algebra and intelligent predictions. What if the same principles were applied by those working to improve outcomes for children?

Steve Aos, keynote speaker at this week's launch of the Working Together for Brighter Futures strategy in Birmingham UK, is an investment adviser to the Washington State legislature of the US. And because children and children's services in Washington do not differ so much from those in other states or in Europe or Australasia, his advice is read by legislators and policy makers around the world as a product guide equivalent to a consumer report.

An economist by training, Steve Aos started his career advising on energy matters. "In the 1980s I was a staffer at an organization called the Pacific North West Planning Council. Legislators wanted to build nuclear power plants. But the US Congress asked them to find out if there was a cheaper way to meet electricity needs. So a group of scientists were commissioned to find out. We built a number of economic models. And we discovered that a variety of measures, such as better insulation and more efficient light bulbs, would lead to cheaper sustainable power without nuclear plants."

In the 1990s, Aos turned his attention to social programs and expected to find similar economic models – and similar respect for evidence. "But I found neither. The models did not exist. Worse, not only was there no evidence base but also the results of science were used inappropriately.

“In the energy field our evidence was rigorously cross-examined by experts and counsel. In the social field, there appeared to be few standards or challenges."

His change of direction was made possible by Washington State Institute for Public Policy which he joined in 1995. Almost unique in the US context, the Institute was established and is supported by the Washington State legislature. It is a non-partisan organization which – as Aos proudly proclaims – is actually perceived as being non-partisan.

"Our Board is a mixture of Democrats and Republicans from the legislature, the executive and other non-partisan organizations. Of course, there are politicians on both sides who believe they know exactly what needs to be done – and don't need any help from us. But there’s a bunch in the middle who welcome a little help in making the right decisions."

People and communication skills are a big part of the job. As he explains, "like any good investment adviser, half of my job is to be a good scientist and half is to help people make sense of the results. The daily walk from home to my office in Olympia, Washington, takes me past the capitol building. The ten minutes here and ten minutes there spent chatting to people, answering questions, starting to understand others are essential to the success of our work."

His achievements have included the systematic reviews of 571 rigorous, real world evaluations of adult and juvenile corrections, interventions and prevention programs from around the world. He has carefully modeled the economics of the various options, including benefits to taxpayers and crime victims and costs to the people of Washington. Based on that understanding he has been able to assemble alternative portfolios and show how they would affect the prison population, and those arguments have helped the state invest in prevention programs and disinvest in prison building.

"Prisons work,” he explains. “If 100% of people were locked up we would have no crime. If we had no prisons crime would be higher. But between zero and 100 per cent is a curve which, like any business, shows diminishing returns for each new prison bed added.

“For the last 35 years in the US we have kept on building prisons. But the impact on crime has become progressively less. My job was to find out the most cost efficient way of helping legislators be tough on crime. As it turned out, prevention programs are a big part of the answer."

It has not all been plain sailing for the Washington Institute. For example, in the late 1990s his team learned the hard way that badly implemented programs do not help and sometimes hinder efforts to improve child well-being. They had begun to implement two programs, Aggression Replacement Training (ART) and Functional Family Therapy (FFT) which they had reason to expect would improve child outcomes. But the results confounded the predictions: they were having no effect.

"It was the 'on average' bit of this finding that was misleading," Aos explains. "When we undertook blind reviews of the therapists working on both programs, we found that those working by the book got the expected effect sizes, or better. But those doing their own thing had poor results. Add the disciplined and laissez-faire therapists together and you had a net effect of zero. But get the programs implemented as intended, and the picture was much rosier."

The discovery led the State government to tie future funding of the programs to a quality assurance scheme. Today, therapists are independently vetted. Those who do not achieve fidelity are retrained or reassigned.

It remains to be seen whether Aos's influence will extend to the UK. This week he gives papers in Birmingham and to a group of senior civil servants from the Westminster government. Either way, Prevention Action will be closely following his progress.