Designing Personal Income Tax and Transfer Reforms: Alternative Modelling Approaches

• Published date: 01 December 2021
• Author: Nazila Alinaghi,John Creedy,Norman Gemmell
• Date: 01 December 2021
• DOI: http://doi.org/10.1111/1467-8462.12416

The Australian Economic Review, vol. 54, no. 4, pp. 445–461DOI: 10.1111/1467-8462.12416

Designing Personal Income Tax and Transfer Reforms:

Alternative Modelling Approaches

Nazila Alinaghi, John Creedy and Norman Gemmell*

Abstract

Decisions regarding personal income tax and

transfer reforms are inevitably subject to value

judgements. Hence, there is a crucial role for

policy analysis in which the implications of

clearly expressed values are considered. Tax

models play a central role in such analyses.

This paper reviewsanumber of tax models and

illustrates their use in the context of the New

Zealand income tax and transfer system. It is

suggested that the appropriate economic

models vary according to: the precise ways

in which the policy question is formulated; the

underlying behavioural responses to taxation

expected across the taxpaying population; the

precise definitions of key variables such as

income inequality; and the specification of

policy objectives such as redistribution,

revenue‐raising or tax efficiency.

1. Introduction

A prerequisite of good tax policy advice is a

rigorous framework within which existing tax

settings and possible reforms can be evaluated.

Tax models play a central role in such exercises.

This paper reviews a number of approaches to

tax modelling that may play a useful role in

designing practical income tax reforms. It

addresses the question of what kinds of informa-

tion can usefully be presented when formal tax

reform proposals are being considered. This

requires clarification of policy objectives and

outcomes so that analyses can shed light on the

merits of alternative reform options.

Especially in the case of personal income

taxes and transfers, which typically aim to

achieve several policy objectives, it is crucial

that reforms are subjected to ‘rational policy

analysis’in which value judgements are

clearly expressed and their implications con-

sidered. For example, instead of simply

making a single recommendation, it can be

important to consider the implications of

adopting a range of explicit value judgements,

thereby presenting a menu of alternative

results as clearly as possible. Different

* Wellington School of Business and Government, Victoria

University of Wellington, New Zealand. Corresponding

author: Alinaghi, email <nazila.alinaghi@vuw.ac.nz>.

We thank two referees for comments on an earlier version

of this paper. We are grateful to the NZ Treasury for

access to the TaxWell microsimulation model, and to

Statistics NZ for access to Integrated Data Infrastructure

(IDI) data. Results reported in this paper are not official

statistics; they have been created for research purposes

from IDI data managed by Statistics New Zealand. The

opinions, findings, recommendations and conclusions

expressed in this paper are those of the authors not

Statistics NZ. Access to the anonymised data used in this

study was provided by Statistics NZ in accordance with

the security and confidentiality provisions of the Statistics

Act 1975. Only people authorised by the Statistics Act

1975 are allowed to see data about a particular person,

household, business or organisation and the results in this

paper have been confidentialised to protect these groups

from identification. Careful consideration has been given

to the privacy, security and confidentiality issues

associated with using administrative and survey data in

the IDI. Further detail can be found in the privacy impact

assessment for the Integrated Data Infrastructure available

from www.stats.govt.nz.

© 2021 The University of Melbourne, Melbourne Institute: Applied Economic & Social Research,

Faculty of Business and Economics

Published by John Wiley & Sons Australia, Ltd

‘consumers’are then equipped to make their

own policy choices.

1

In choosing a framework with which to

evaluate a tax system, or proposed reforms, it

is also important to recognise the large gap

between the simple conceptual models typi-

cally examined within public finance theory,

and the models required for tax policy advice

in practice. While the former can provide

important high‐level insights, concerning, for

example, those taxes that are expected to be

associated with greater behavioural responses

or redistributional properties, they cannot

generally provide robust conclusions in spe-

cific policy contexts. As a result, an important

starting point is discussion of the type of

economic model that can usefully be em-

ployed in a given policy context. Since all

models are subject to limitations, necessarily

involving a number of abstractions from

reality, and are designed for specific purposes,

they cannot be expected to provide informa-

tion about all possible responses to, and

effects of, tax changes. Analyses therefore

usually need to be presented with appropriate

caveats, and guidance over those which are

expected to be most relevant.

Personal income tax is the tax most often

tasked with addressing equity aspects within

the tax system, and is recognised to have

direct efficiency consequences. To highlight a

particular concrete example of design choice,

special attention is given to changes in the top

marginal income tax rate and the income

threshold above which that rate applies. Such

changes have been given considerable atten-

tion in recent tax debates in various countries,

with the appropriate amount of tax levied on

the ‘top 1 per cent’(or top 0.1 per cent) of

taxpayers becoming something of a fixation in

popular inequality debates and related aca-

demic studies.

2

Focus on the top income tax

rate and threshold often derives from the

mistaken view that tax rate progression

(which reflects the extent to which marginal

rates increase as taxable income increases) is

necessary for income redistribution. However,

it is the combined effect of different taxes and

social transfer payments, and their impact on

average tax rates, that ultimately matters for

redistribution, rather than simply the income

tax schedule.

The use of alternative approaches is illu-

strated in the context of the New Zealand

income tax and transfer system, but many of

the modelling and policy choice issues

discussed have much wider applicability.

The NZ system has many features in common

with other OECD countries, and involves a set

of (currently four) marginal income tax rates

that provide rate progression across a set of

rising income thresholds.

3

A distinguishing

feature is that there is no tax‐free income

range, so that the first rate of 10.5 per cent

applies from the first dollar, although various

social welfare rebates also apply.

Nevertheless, like many OECD couurties,

the marginal income tax rates, particularly in

the lower‐income ranges, do not reflect

effective marginal rates in view of the

existence of a range of means‐tested benefits

with various taper or abatement rates.

In addition, New Zealand shares with the

UK and some other countries, a family tax

credit system that pays tax credit to families

with children, based on household income

levels and, in some cases, hours of work.

These refundable tax credits, paid as lump‐

sum transfers even when no taxable income is

earned, serve to reduce the effective average

tax rate for families with children. Indeed,

they generate negative average rates for the

lowest 3 to 4 deciles of the taxable income

distribution.

4

These tax‐transfer complications

provide one reason why carefully designed

microsimulation models are important for

analysis of tax reforms.

The paper is organised as follows. Section 2

discusses the use of two types of microsimu-

lation model –arithmetic and behavioural—to

compare reforms. As mentioned earlier,

emphasis is placed on the effects of an

increase in the top marginal tax rate. An

advantage of microsimulation models is that

they enable detailed examination of the

potential effects of specified tax and transfer

changes, and a wide range of evaluation

methods, thereby allowing people to assess

outcomes based on their own value judge-

ments. Section 3 demonstrates how a

446The Australian Economic ReviewDecember 2021

© 2021 The University of Melbourne, Melbourne Institute: Applied Economic & Social Research,

Faculty of Business and Economics