The Labour Market Effects of ‘Working for Families’ In New Zealand

• Author: John Creedy,Joseph Mercante,Penny Mok
• DOI: http://doi.org/10.1111/1467-8462.12260
• Date: 01 June 2018
• Published date: 01 June 2018

The Labour Market Effects of ‘Working for Families’

In New Zealand

John Creedy, Joseph Mercante and Penny Mok*

Abstract

This article examines, using a behavioural

microsimulation model for New Zealand, the

impact of the major package of welfare reforms

collectively referred to as ‘Working for Fami-

lies’(WfF). These reforms were announced in

May 2004 and involved a phased introduction

of changes, with annual modif‌ications from

October 2004 until April 2007. Concentrating

on welfare payments for families with depen-

dent children, this remains a cornerstone of the

New Zealand welfare system. The simulation

takes a backward-looking perspective, by

comparing labour supply under the 2008

database and tax structure with that obtained

using the 2008 data and 2004 tax structure

(suitably adjusted for inf‌lation).

1. Introduction

The aim of this article is to examine, using

a behavioural microsimulation model for

New Zealand, the labour market impacts of

the major package of welfare reforms collec-

tively referred to as Working for Families

(WfF). These reforms were announced in

May 2004, and involved a phased introduction

of changes, with annual modif‌ications from

October 2004 until April 2007. Concentrating

on welfare payments for families with depen-

dent children, WfF remains a cornerstone of

the New Zealand welfare system. Despite the

lapse of time since the full implementation of

the new structure, a simulation analysis of the

implications for labour supply is warranted in

view of its continued importance and the fact

that very little evidence is actually available

about its effects.

1

The two aims of thereforms, explicitly stated

at the time, were to reduce child poverty and

to improve work incentives for low-paid work-

ers. A main labour supply stimulus involved

the use of an in-work payment for families

with children, available after a threshold

number of hours of work. For couples, the

hours threshold is 30 hours per week, and for

sole parents the threshold is 20 hours per week.

In addition, changesto abatement rates (applied

above earnings thresholds) were made in

order to reduce effective marginal tax rates

facing families with children. In New Zealand,

these rates, variously termed ‘clawback’or

‘taper’rates elsewhere, are referred to as

‘abatement rates’.

Despite the emphasis on stimulating labour

supply, and the complexity in the budget

constraints facing individuals, the government

was not able, during the lengthy planning

* Creedy: Victoria University of Wellington, Wellington,

New Zealand; Mercante: Australian Department of

Employment, Canberra, Australia; Mok: Ministry of

Business, Innovation and Employment, Wellington,

New Zealand. Corresponding author: Creedy, email

<john.Creedy@vuw.ac.nz>. This article is part of a larger

project on ‘Improving New Zealand’s Tax Policy via

International Tax Transfer Model Benchmarking’, funded

by an Endeavour Research Grant from the Ministry of

Business, Innovation and Employment (MBIE) and

awarded to the Chair of Public Finance (CPF). Access to

the data used in this article was provided by Statistics

New Zealand in accordance with security and conf‌identi-

ality provisions of the Statistics Act 1975. The results

presented in this study are the work of the authors, not

Statistics New Zealand. The authors are grateful to

Matthew Bell, Guyonne Kalb and two referees for their

very helpful comments on an earlier draft.

The Australian Economic Review, vol. 51, no. 2, pp. 211–31

°

C2018 The University of Melbourne, Melbourne Institute of Applied Economic and Social Research

Published by John Wiley & Sons Australia, Ltd

stages of the reforms, to draw on advice based

on labour supply modelling. Reliance was

placed on the argument that the introduction of

an in-work payment, involving the hours

thresholds, would provide a strong incentive

to increase hours of work. During the planning

stages, the program was referred to as ‘Future

Directions’; background details can be found in

Ministry of Social Development (2004a,

2004b) and Nolan (2002).

The plans were signif‌icantly inf‌luenced

by experience with tax credits in the UK,

discussed, for example, by Brewer (2003), and

the labour supply estimates of Duncan and

Giles (1996) and Blundell et al. (2000). It was

also possible to draw on studies of experience

with US tax credits, reported for example in

Eissa and Hoynes (1999) and Ellwood (2000).

Potential incentive effects were discussed by

Nolan (2003) and a theoretical analysis of the

role of hours thresholds with in-work tax

credits, along with simulations in the optimal

tax tradition, motivated by the WfF reforms,

was reported in Creedy (2005). A useful wide-

ranging early review is by Johnson (2005), who

compared WfF with US social transfers and

stressed a number of adverse incentive effects.

The eligibility rules, the interaction with

other aspects of the tax and transfer system, and

unique features of the New Zealand system

mean that the impacts on work incentives

are not straightforward. The changes actually

resulted in higher effective marginal tax rates

for some families. In addition, the reforms

generated both income and substitution effects,

so that labour supply could either increase or

decrease, and the effects could differ for

different demographic groups. Hence, it is

important to understand the impacts of the

reform on potential labour market behaviour of

the targeted groups using a behavioural micro-

simulation model that captures the full hetero-

geneity of the population as well as the full

complexity of the income tax and transfer

system.

Two earlier studies of the labour market

effects of WfF have been carried out. The

f‌irst was reported in an unpublished paper

by Kalb et al. (2005), which was later

included in Buddelmeyer et al. (2007).

2

This

used the behavioural microsimulation model,

TaxMod-B, and involved comparisons of

labour supplies under the 2001 tax structure

and Household Economic Survey (HES) with

those simulated using the planned 2008 tax and

welfare structure.

3

The 2001 wage rates, non-

wage incomes and benef‌it levels were uprated

to their projected 2008 values.

4

The base

year of 2001 was used because it was the

most recent HES, following the Statistics

New Zealand decision to reduce the frequency

of the survey. The survey weights were

recalibrated in order to produce a sample that

more closely approximated that expected in

the later implementation years.

5

The second study was an ex post evaluation

of the reform, carried out jointly by the

Ministry of Social Development (MSD) and

the Inland Revenue Department (IRD). It

covered the period from implementation of

the scheme until the end of March 2007.

Dalgety et al. (2010) estimated gains of six

percentage points in employment rates for sole

parents.

6

They also found that both members

of couples with children were less likely to be

in employment after the reform, by around

two percentage points. In addition, WfF was

estimated to have decreased the employment

rate of secondary earners, with the largest

impact on secondary earners working part-time

(fewer than 30 hours per week). The evaluation

reports identif‌ied primary earners in couples as

those with greatest attachment to the labour

market, being predominantly male, having

higher qualif‌ications and being slightly older

than secondary earners. In using the difference-

in-differences approach, Dalgety et al. (2010)

relied on the very strong assumption that sole

parents (the comparison group) are similar to

single adults without children (the control

group), except that sole parents are affected by

the WfF reforms.

7

They also estimated changes

in the hazard function, using a dummy variable

for the post-reform period. Hence no explicit

labour supply modelling was used.

Importantly, numerous macroeconomic and

other changes took place over the period

following the initial introduction of WfF,

in addition to those made to the tax and

welfare system. As a result, direct comparisons

212 The Australian Economic Review June 2018

°

C2018 The University of Melbourne, Melbourne Institute of Applied Economic and Social Research