Whose research to believe?

On the declining teen birthrate:

The main direct factor in recent declines has been the increased use of contraception, but the other is delaying age of first sexual experience and the nature of that experience.

Underlying differences in the use of contraception and age at first sexual experience are a range of social, economic and cultural factors. Socio-economic disadvantage is a key factor, and this is broader than financial disadvantage: it involves broader social capital factors such as parental involvement, education, employment opportunities, leisure activities and community involvement.

That's according to the latest NZ research  commissioned by Superu and produced by the University of Waikato.

But the NZ Herald reports today:

A paper called Is Education the Best Contraception? The Case of Teenage Pregnancy in England, published in Social Science and Medicine, said: "Educational performance is significantly associated with lower teenage pregnancies and the estimated effects are large. For example, a 10 per cent increase in GCSE [end of school results] implies a reduction in the teenage conception rate of about 8 per cent."

Figures released last month showed the pregnancy rate for under-18s in England - although still among the highest in the west - has fallen by more than 40 per cent.
"Given that the numbers achieving five good GCSEs have increased by about 50 per cent since 2004, this factor alone has the potential to explain a large proportion of the recent decrease," the paper said.

Previously it was thought that long-acting, reversible forms of contraception (Larcs), such as IUDs, were a key factor. But the study found a large increase of the use of Larcs had only a small effect. A 10 per cent increase in their use was associated with a reduction in the under-18 conception rate of about 0.3 per cent.

The NZ finding is largely based on literature reviews from overseas countries. For instance:

Kearney and Levine (2012) compared the United States to countries with lower teenage birth rates and found the main difference was in the extent of contraception use rather than sexual activity. US teenager s had a lower level of sexual activity than many of the European countries with lower birth rates, but also had a much lower level of contraception use. However, both sexual activity and contraception use are moving in positive directions with sexual activity declining and contraception use increasing, thus both contributing to the decline in teenage births in the United States .

Are the welfare reforms given any credit for declining teenage birth rates? Again from the NZ literature review:

Levine (2002) and Hao et al (2007) found no relationship between teen

age pregnancy and welfare policies in general. But Hao et al (2007)

did find that strong enforcement of child support payment policies for fathers deters teenage pregnancies and births and increases school enrolment of teenage girls. Horton (2006) found a small decline (3.25% to 2.8%) in teenage childbearing following the 1997 US welfare reform aimed at making assistance temporary and encouraging work.

The NZ paper goes on:

Increased contraception use and reduced welfare benefits have been used to explain the first period of decline in the United States. But the role of policy is likely to be secondary as a similar decline has occurred in most other developed countries.

Except "most other developed countries" have been reforming their welfare systems similarly - the UK, Australia, Canada and NZ.

Here's the only thing I can say for sure.

If you have an opinion on why the teenage birth rate is declining - for example, welfare reform as a significant factor - it won't take you long to find a paper that supports it. And another that rejects it.