Image by Adam Selwood CC 2.0 via Flickr

The Treasury recently released CAPITA, a model of Australia’s personal income tax and transfer system. CAPITA is a key tool used by government to analyse the impact of policy reforms on households, and it has been released to the public to allow more people to contribute rigorous, evidence-based policy analysis to Australia’s public policy debate. CAPITA was developed jointly by the Treasury, the Department of Social Services and the Department of Employment, with assistance from the Australian Bureau of Statistics and the Parliamentary Budget Office.

CAPITA is a general-purpose static microsimulation model of Australia’s personal income tax and transfer system designed to analyse the distributional impacts of policy across the Australian population and highlight interactions between the tax and transfer systems. Figure 1 shows the key components of CAPITA.

The basefile contains detailed unit-record data on incomes and demographic characteristics used to represent the modelled population. The policy calculator consists of a set of rules for calculating the taxes and transfers applicable to each individual. Together, these components deliver the tax-transfer outcomes of the individual units, which can be aggregated as desired. The rules in the policy calculator can be varied to examine the effects of changes to personal income tax and transfer policies on individuals, families and groups within the population.

Figure 1: Components of CAPITA

As part of its suite of models, Treasury has used static microsimulation models for over 20 years to analyse the immediate distributional effects of changes to tax and transfer policy. Previously, Treasury used STINMOD (Static Incomes Model), a tax-transfer microsimulation model developed by the National Centre for Social and Economic Modelling.

Static microsimulation models provide an estimate of short-run effects for policy changes that do not change economic structure significantly. However, as static models assume no behavioural change in response to policy change, they do not estimate long-term effects of policy changes or the effects of policy changes that are likely to change economic agents’ behaviour.

How CAPITA models policy impacts?

Going into more detail on the model, the initial CAPITA basefile is based on the Australian Bureau of Statistics’ Survey of Income and Housing (SIH), which collects detailed demographic and income information from a sample of households. In 2013-14, 14,162 households were surveyed, representing 97 per cent of the population. Each household is assigned a weight to indicate the number of households it represents to ensure the survey data is representative of the population of interest.

To model the effect of policy changes in future years, the characteristics of the individual units are held constant over time and the original snapshot is ‘aged’. This ageing process involves adjusting weights of the individual units, so that the population matches broad demographic trends, and inflating economic data, such as incomes, using an appropriate inflator.

In the policy calculator, CAPITA models the major payments in the transfer system including pensions, allowances and family payments and the significant structural elements of the personal income tax system including marginal tax rates, the Medicare levy and major tax offsets. The payments included were selected by considering the fiscal size of the policy and the number of people affected. The trade-off between coverage of the transfer system and complexity of the resulting code was also taken into account.

To complement the distributional analysis, CAPITA can also conduct ‘cameo’ analysis by calculating the tax-transfer outcomes for a hypothetical family as a function of their income. This is used to provide examples of the impact of policy and policy changes. The basefile for cameos is a synthetic dataset of user-specified family attributes and income levels. The tax-transfer calculator generates the outcomes for each hypothetical family.

CAPITA can also calculate Effective Marginal Tax Rates (EMTRs) for hypothetical individuals and families. EMTRs refer to the proportion of each additional dollar earned that is removed due to either personal income tax paid or the withdrawal of transfer payments. EMTRs are useful from a policy perspective in terms of assessing potential financial incentives arising from the design and structure of the tax and transfer system, which (among other things) can be a factor that people consider when deciding when and how much they wish to work.

Further details on the model can be found in a paper describing CAPITA on the Treasury Research Institute website and the documentation released alongside the SAS source code. Users of the model will require access to the Australian Bureau of Statistics data underpinning it and must comply with the CAPITA user licence agreement. Treasury would like to collaborate with model users to improve and develop CAPITA into the future. Users are invited and encouraged to share any improvements or suggestions. Importantly, users should note the Australian Treasury does not take responsibility for any results generated by the model.

This article has 2 comments

  1. Pingback: Open Budget Survey 2017 Part 2: What Can Australia Do to Improve our Budget Process? - Austaxpolicy: The Tax and Transfer Policy Blog

  2. Pingback: Budget Forum 2018: A Missed Opportunity for Enhancing Australia’s Budget Transparency on Distributional Information - Austaxpolicy: The Tax and Transfer Policy Blog

Leave a comment

Your email address will not be published. Required fields are marked *