Behavioural Effects of Pension Eligibility in New Zealand

Abstract

This study examines the differential labour supply behaviour of pension eligibility in New Zealand, including New Zealand Superannuation (NZS) take-up rates, the composition and level of personal income, and individual employment rates. Administrative data from the Integrated Data Infrastructure (IDI) covering the period from five years before to five years after the pension age was used to examine cohorts born between 1940 and 1947. Using Ordinary Least Squares (OLS), maximum likelihood Probit regression, and hazard-based duration models, a variety of statistically significant labour supply effects were discovered.

First, a 96.4 percent of NZS take-up rate was found three months after turning 65 for the entire sample cohort that was restricted to individuals who were presumably eligible for this public pension at 65, and it continued to increase over time, reaching 98.7 percent at age 70. The NZS take-up rate gaps for various demographic groups narrow over time. Some take-up issues appear to exist with NZS, which appears to be concentrated among people of Māori, Pacifica, and Asian, as well as people who had continuously received main benefits between the ages of 60 and 64.

Second, a temporary boost of $253 in total income on average was found for the entire sample cohort at the 65th birthday month, due to the effect of receiving NZS. A permanent rise of $565 in total monthly income was discovered after being pension eligible, which was equivalent to a substantial 30.8 percent of the average income at the age of 64. Furthermore, we found that NZS was not entirely offset by a reduction in earned income at age 65. Instead, NZS generated a substantial boost to total income when compared to income prior to eligibility and lasted for nearly five years for the entire sample cohort. Only at age 69.44, was average income predicted to be at the same level as that received in the month immediately before becoming pension eligible.

Third, being pension eligible significantly lowered the employment rate by 3.38 percentage points for the entire sample cohort, which we suspect was due entirely to the pure income effect generated by NZS. For the sub-sample who transitioned from means-tested main benefits to the non-means-tested NZS with weaker work disincentives, there were significant positive relative gains in the labour supply. These effects varied based on the uptake of main benefits between the ages of 60 and 64, with people who had continuously received main benefits having a greater impact.