The Impacts of Intertemporal Income Shifting on Estimates of the Behavioural Response to a Tax Reform: Evidence from New Zealand

Abstract

On 7 December 2020, the New Zealand Parliament added a new 39 percent top marginal tax rate on personal income over $180,000. The aim of the policy was to increase tax revenue gathered from the top two percent of income earners. The introduction of the new increased top marginal tax rate changed the labour-leisure trade-off for New Zealand’s top earners. In this thesis I estimate the so-called Elasticity of Taxable Income (ETI). Using the standard specification, I find an elasticity of 0.227 which is in line with contemporary New Zealand estimates. The eve of a tax reform creates a two-period discontinuity with a clear tax preferred side. This creates a strong incentive for intertemporal manipulation of income towards the lower marginal tax rate period. I show evidence of income acceleration, likely in the form of bonus payments and other discretionary payments being paid in March 2021, the preferred tax period. I find evidence that approximately 4.2 percent of taxpayers engaged in accelerating behaviour. This has implications for estimation of the ETI. Using a specification that removes the months with the highest level of income acceleration I find an estimate of the ETI of 0.139. The results are based on full population monthly wage and salary data provided to the Statistics New Zealand Integrated Data Infrastructure (IDI) by Inland Revenue New Zealand.