Here at JTf we try to bring attention to as much of the latest reputable education policy research as possible. You can check our scribd account for the back catalogue. What consistently comes to light are the following observations

Education policy covers a vast area of issues
A lot of research is paid for by organizations with an agenda
Most research is in a state of relative infancy
There's a lot of contradictory research

As we have been delving into one policy area, namely teacher attrition, we came across 2 studies. The first form the National Center for Education Statistic, titled "Beginning Teacher Attrition and Mobility: Results From the First Through Third Waves of the 2007–08". One of the questions this study is looking at is the impact of early career mentoring and its impact on attrition. This study has tentatively found

Among beginning public school teachers who were assigned a mentor in 2007-08, about 8 percent were not teaching in 2008–09 and 10 percent were not teaching in 2009–10. In contrast, among the beginning public school teachers who were not assigned a mentor in 2007–08, about 16 percent were not teaching in 2008–09 and 23 percent were not teaching in 2009–10 (table 2).

In other words, having a senior teacher mentor a younger teacher has quite a dramatic effect on the attrition rates of new teachers. This is an important finding for many reasons, but especially important to understand if we were to move away from collaborative work places to a more merit based system where competition rules.

A second study, from the US Department of Education titled "Impacts of Comprehensive Teacher Induction Final Results from a Randomized Controlled Study", had a different finding regarding the question of mentoring

Neither exposure to one year nor exposure to two years of comprehensive induction had a positive impact on retention or other teacher workforce outcomes
[...]
-There was no impact on teacher retention over the first four years of the teachers’ careers. This was true of retention in the original school, the original school district, and the teaching profession.

These are the kinds of problems that need to be wrestled with, in order to produce the most effective education system possible.

For the record, mentoring does produce important positive results, as the Dept. of Ed study found

• For teachers who received two years of comprehensive induction, there was no impact on student achievement in the first two years. In the third year, there was a positive and statistically significant impact on student achievement.

- In the third year, in districts and grades in which students’ test scores from the current and prior year are available, students of treatment teachers outperformed students of the corresponding control teachers on average. These impacts are equivalent to effect sizes of 0.11 in reading and 0.20 in math, which is enough to move the average student from the 50th percentile up 4 percentile points in reading and 8 percentile points in math.

- These results are based on the subset of data for which students’ test scores from the current and prior year are available. If the analyses are conducted without requiring test scores from the prior year, we do not find an impact on math or reading scores. This alternative approach nearly doubles the available sample of study teachers but the lack of data on students’ prior achievement results in a less precise estimate. This means that we are less likely to detect a true impact if it exists, despite the larger sample size.

We can't build a strong education system without experienced educators prepared to mentor young teachers, and that mentorship requires multiple years of effort to show results.