The extra two deaths of babies per 1000 born was high, Ms Lawton said, and statistically and clinically significant.
The study, which also featured work by researchers in Otago and Illinois, was published today in the International Journal of Gynecology and Obstetrics.
It also found a positive trend for increasing safety as midwives' years of experience increased.
For midwives who were nurses first, no association between experience and perinatal mortality was found.
The study's findings have prompted Ms Lawton to call on the Government to urgently review New Zealand's maternity training system.
She wants it to consider implementing mandatory clinical supervision for newly-graduated midwives for up to two years in hospital followed by a supervised year in the community for independent midwives.
Ms Lawton stressed that the study pointed to a training issue for New Zealand midwives and the need for a system change to reduce unsupervised practice for junior midwives.
"It is not about midwives as practitioners," she said.
"Midwives are expert clinicians giving quality care and are the backbone of the New Zealand maternity system.
"However, this finding for first-year direct-entry midwives is concerning and we need to be prepared to critique the training, and ensure we have a maternity system where mothers can have faith that the system they trust is the best it can possibly be."
In New Zealand, there is no requirement for supervision in the first years after graduation for midwife-only midwives, although a mentorship programme was introduced from 2011.
Due to well-publicised deaths of babies, the mentorship programme has been strengthened.
But the researchers believe that until this is changed to supervision the measure is unlikely to make a substantial difference.
Australia has introduced an "eligible midwife registration standard" that requires the midwife to complete the equivalent of three years full-time post registration experience as a midwife and give evidence of current competence to provide pregnancy, labour, birth and postnatal care.
Earlier this year, Coroner Gary Evans made strong recommendations for changes in the teaching and supervision of graduate midwives in his findings on the deaths of a mother and newborn.
Regulatory body the Midwifery Council today questioned the validity of the research.
Council chairwoman Associate Professor Dr Judith McAra Couper said the research had "limitations" and raised issues over the data's "accuracy and age".
"The authors of the study themselves state there is no causality between new midwives and increased perinatal mortality," she said.
"So while one thing could be related to another it does not tell us that one caused the other. The new midwives could have looked after more women with high risk conditions - those who had premature babies, were smokers or had increased BMI.
"By not taking these things into account it is hard to know what significance or meaning this research has."
The council also raised concern that the data was 6-10 years old and did not take into account the "many changes" that have taken place in midwifery education since 2009.
A review of the Midwifery First Year of Practice (MYFP) was undertaken in 2012-2014 by the council, Ministry of Health, health officials, and the College of Midwives.
"At this time the question of new graduate midwives going into hospital was carefully scrutinised," Dr McAra Couper said.
"There was no evidence at all to show that this was the best and safest option for mothers and babies - instead the evidence showed the best option was to strengthen the mandatory first year of practice programme."
Dr McAra Couper said that while the council will "closely examine this research" at its next meeting, women and their families should be assured that midwives are "providing a high quality and safe service".
The New Zealand College of Midwives also blasted the study for "poor data quality and flawed methodology".
"Dr Lawton has mixed up information," college chief executive Karen Guilliland said.
"For example, this study assumes the midwife initially registered to work alongside a woman is the same midwife with the woman at the birth and that is not correct.
"As a pregnancy progresses and risk factors are recognised by midwives, women are referred or transferred to hospitals and the caregiver can change."
She added that since the study's data period, the rate of baby deaths during labour has significantly reduced, with 2015 currently sitting at an all-time low.
"Clearly the presence of highly educated midwives has improved the birth outcomes for babies. This is in contrast to the misguided claims made by Dr Lawton's study," said Ms Guilliland.
