The paper is here; Milligan also discusses the results in Canada's Globe and Mail.
We look at children born in the years just before and just after the 2001 expansion of paid parental leave through the Employment Insurance system from 25 weeks to 50 weeks. Looking across various measures of child temperament, physical development, and child security, we find very little evidence of an impact of this large expansion of parental leave entitlements. As one example, the pre-reform average age at which the child took his or her first steps was 9.46 months. Our study finds that this changed by only 0.06 months after the reform -- and that result is not distinguishable from zero. Our work in this study builds on previous studies we have written on the labour-market implications of parental leave and on the health implications for children.New Zealand's Labour Party has been making noises about extending paid parental leave. The Canadian evidence suggests it doesn't provide that much benefit. And, I'd worry a fair bit that the more employers expect that they'll have to fill holes with temporary positions, the less likely they'll be to hire women they think likely soon to have another kid.