HMRC/HMT/ESRC Joint Research Programme

Tax Policy and Operations in the Context of Economic and Societal Change

About 18 months ago the ESRC/HMRC/HMT jointly funded seven research projects on tax administration. Gareth Miles,1 reports on the results.

Other items related to this theme, published in the Newsletter, include:
Mirrlees Review (January 2013)
Taxation in a global economy (January 2014)

In 2010 the Economic and Social Research Council, HM Revenue and Customs, and HM Treasury jointly issued a call for research projects “focusing on tax policy and operations covering business and personal tax, benefits and credits, compliance and enforcement”. The call indentified several areas as important for research. These included improved understanding and modelling of behavioural impacts on tax policies, and developing better approaches to forecasting and modelling of the effects of tax structures and tax reforms. Seven projects were eventually chosen to receive financial support and papers from three of the projects were presented at the 2012 Royal Economic Society conference held in Cambridge. These papers gave a good insight into the variety and quality of the research funded by the joint ESRC/HMRC/HMT programme.

Richard Disney’s project “Household responses to complex tax initiatives” involves researchers from the Institute for Fiscal Studies and the University of Nottingham. The project examines how tax schedules give strong incentives for households to arrange their tax affairs so as to minimise their tax burden, but introduce complexities that forestall full optimisation. Consequently, individuals and households may respond to tax regimes and to tax reforms by approximations (for example, estimating marginal tax rates off estimates of annual earnings rather than total income in the tax year, or by ‘spotlighting’ certain features of the tax regime and ignoring other components). Using both econometric analysis and a laboratory experimental, the research investigates whether individuals and households respond to non-linear tax structures and to tax reforms that change incentives, through retirement saving and labour supply decisions.

In a paper presented at RES 2012 stemming from this research project, Rowena Crawford, Richard Disney, and Carl Emmerson looked at the effect of marginal tax rates on retirement saving. Given that tax reliefs on retirement saving are given at the marginal rate, there is obviously a stronger incentive for individuals on higher rates of income tax to engage in greater retirement saving. Less well understood is that individuals in receipt of tax credits could also engage in retirement saving as a means of increasing their eligibility for tax credits, since these are assessed on income net of pension contributions. Finally, couples can optimise their tax affairs by the individual with the higher marginal rate taking on the bulk of pension contributions. The research examined all these possibilities.

Given the importance of these policy issues, it is something of a surprise to find that there is little previous literature in Britain on these policy concerns. One reason is the lack of good data on consumption and saving matched to asset holdings and pension contributions with which to pin down the effects of tax schedules and policy reforms. In the context of income tax rates, a further issue is that, while it is well-established that individual with high average tax rates save a greater share of income, this may simply reflect differences in time preference and access to employer-provided pension plans rather than the impact of the tax regime per se.

The authors therefore looked for discontinuities in saving behaviour at marginal tax rate thresholds. This involved computing original (i.e. pre-tax and pre-pension contribution) income and earnings relative to taxable income and then looking for changes in behaviour across households at changes in predicted marginal tax rates. An example of the effect is given in Figure 1: it should be noted that the (relatively small) discontinuity in the probability of contributing to a pension around the threshold of the 40% tax rate appears to take place based only on a measure of earnings rather than full income; this may simply reflect a lack of knowledge of full income in the year until the end of the tax year, or approximating behaviour among households.

When investigating joint tax decisions, the research found that, if anything, the probability of contributing to a pension increases among both partners when one or other is at the high rate threshold – this may reflect some form of ‘signal’ by which a high rate of tax on one or other family member induces both members to plan their saving strategy more carefully – bearing in mind that there is an incentive to reduce tax liabilities by retirement saving at any positive rate of income tax. Finally, there the research found no evidence that people understand the interaction between tax credit eligibility and tax reliefs on retirement saving, although the lack of response may also reflect liquidity

The second RES 2012 paper from the Joint Research Programme summarised results from the project “A lifetime perspective on the distributional and incentive effects of the tax system” led by Monica Costa Dias from the Institute for Fiscal Studies. The project seeks to increase understanding of the mechanisms driving individuals' decisions to study and to work, and how the tax and benefit system influences these decisions. It is distinguished from previous contributions through its focus on the intertemporal dimension that is involved in many of the important decisions that people make during their lifetimes. It aims to answer questions such as: how financial incentives to work change over the lifecycle; how the tax burden shared over the population and over the lifecycle; and what impact would changes to taxes and benefits have on the distribution of lifetime income and the incentives to work and invest in human capital?

The paper presented at the RES conference was joint work with Richard Blundell, Costas Meghir, and Jonathan Shaw. It considered the long-term effects of in-work benefits in a life-cycle model designed for policy evaluation. In-work benefits have gained attention over the past 20 years as a number of countries, including the US, Canada, the UK and France, have allocated gradually increasing resources to such schemes. The basis of in-work benefits are that they provide a method of transferring income towards low-income families provided that members of the family are in work. The schemes are generally implemented as a subsidy to working and frequently dependent on family composition, particularly the number of children. Their objective is to alleviate poverty whilst simultaneously mitigating some of the adverse effects other benefits have on the incentives to work.

The basis of the paper is that the generosity of in-work benefits may affect life-cycle decisions other than just employment. For example, the value of additional education may be affected by an increase in the value of employment through the insurance effect of in-work benefits. The insurance effect may be substantial, with partial protection against bad income shocks, which encourages individuals to remain in work and increase labour market attachment. This can have dynamic consequences on future employment and earnings through the effect on work choices and experience. These dynamic links may be of great importance in welfare evaluation. Responses in anticipation of being (directly) affected by a policy in the future may enhance its effects. For example, education decisions may be responsive to an expected change in the return to education caused by future policy.

The paper analysed these effects in a life-cycle model of women's labour supply which incorporated human capital formation. Household decisions were taken in an uncertain dynamic environment with constraints on the availability of credit. The model included important features not previously considered in a single model. First, the dynamic process of family formation was explicitly accounted for. Second, a detailed description of the policy environment was used to accurately determine net earnings by employment status. The model was applied to assess the impact of UK in-work benefits. In-work (i.e. work-contingent) benefits were first introduced in the UK in 1971 with the Family Income Supplement. Several changes have occurred over time, with the scheme being re-labelled as Family Credit in 1988 and Working Families' Tax Credit (WFTC) in 1999, and then split into the Child Tax Credit (CTC) and Working Tax Credit (WTC) in 2003 (Child Tax Credit can be claimed by those not working). Generosity has generally increased over time, with the value of awards increasing and eligibility extended to more families, at least partly through reductions in the rate at which awards were tapered away.

The model was calibrated on a large set of data moments from the nineties under a policy environment reproducing the April 1999 regime. Many important empirical features were closely reproduced, including the empirically estimated short-run effects of the WFTC reform on employment rates. Simulations were then used to study the impact of WFTC and WTC/CTC on women employment, family income, and education decisions. The results show a small but non-negligible anticipation effects on employment and education. There was a substantial employment effect for lone mothers and mothers in couples but a small impact on education choices. The employment effect was due mainly to reallocation of hours across the lifecylce rather than changes in the nature of employment. The results identified some anticipation effect, but this had little impact on employment during eligibility.

The third of the Joint Research Programme papers presented at RES 2012 was “Distortionary Taxation, Debt, and Immigration” Michael Ben-Gad from City University addressed the extent to which governments can shift the cost of government expenditure from today's voters to future generations of immigrants without resorting to taxation that is explicitly discriminatory between the groups. A government that can shift the cost onto future immigrants is able to buy current popularity by financing current expenditure without placing the burden of debt service on current voters or their descendants.

The analysis investigated the behaviour of an optimal growth model with overlapping dynasties. This allowed quantification of how much the government could shift the burden of taxation between the present-day inhabitants of a country and the future immigrants. The policy tools at the disposal of the government were the rates of distortionary factor taxation and level of deficit finance. The best perspective from which to judge the finding of the paper is relative to the results for a standard macroeconomic model with representative agents and inelastic labour supply. In such a model, if government expenditure is a fixed share of net national product, the government can minimize the deadweight loss of factor taxation by holding the tax rate on capital constant over time and equalizing the long-run tax rates on capital and labour income. Furthermore, fluctuations in the tax rate on labour income are neutral and debt neutrality applies so the distribution of the debt burden over time does not matter. Hence, tax-smoothing is optimal in the long-run.

In contrast, in a model of an economy without representative agents, in this case one that is absorbing a continuous stream of immigrants, debt is not neutral. This creates the potential for redistributing resources from future immigrants to natives by deviating from standard optimal policy, and the immediate welfare gains from doing so can outweigh the loss in efficiency. This effect generates a bias in favour of deficit finance. The bias in favour of deficit finance is not without limit: today's population or their descendents must still pay higher taxes and live with the distortions these taxes generate along with the immigrants. Instead, for a given rate of immigration and policy horizon, the government will choose policies that balance the deadweight losses associated with fluctuating tax rates suffered by the initial population and its descendents, against the benefits that accrue to this same population if postponing part of their tax payments can shift some part of the burden to future arrivals. This can be to the benefit of people already resident in the country and their descendents.

The flow of most international migration is from low-income to high-income countries so migrants typically arrive at their destination with negligible amounts of capital, at least in comparison to the stock of capital owned by the typical established resident. This provides a explanation for why the government might choose to shift the burden of taxation towards wages from which immigrants derive disproportionate share of their income and lower the tax on capital while maintaining a balanced budget.

Given this theoretical finding the talk proceeded to test whether the effect could apply for realistic parameter values. Michael used data for the United States between 1980-2010 to calibrate his model and calculate the optimal fiscal policy from the perspective of the native population. This was undertaken for a wide range of different rates of immigration. The welfare gain that accrued to the members of the native population as policymakers shift both the degree to which they rely on deficit finance to pay for government expenditure and the length of time during which they continue to accumulate this extra debt were quantified. The calculations showed that the predictions of the model did a good job of matching the forecasts produced by the US Congressional Budget Office under their Alternative Fiscal Scenario which predicts rising levels of the debt burden. The paper showed how expenditure policy can be influenced by the presence of immigration and gave a new perspective on rising debt levels.

The three papers from the Joint Research Programme presented at RES 2012 demonstrate the benefits of collaborative funding between ESRC and HMRC/HMT. The joint funding model has generated research of the highest academic quality but with relevance to strategic policy-making.

The seven projects financed by the Joint Research Programme were:
“A lifetime perspective on the distributional and incentive effects of the tax system”, Principal Investigator: Monica Dias, Institute for Fiscal Studies.
“Company births and deaths: investigating the role of taxation”, Principal Investigator: Michael Devereux, University of Oxford.
“Distortionary Taxation, Debt, and Immigration”, Principal Investigator: Michael Ben Gad, City University.
“Household responses to complex tax initiatives”, Principal Investigator: Richard Disney, University of Nottingham.
“Optimal Audit Policy for a Tax Evasion Network”, Principal Investigator: Gareth Myles, University of Exeter.
“The Effects of Taxation on Heterogeneous Firms: Micro Level Evidence Across Europe”, Principal Investigator: Richard Kneller, University of Nottingham.
“The influence of decision making costs on the effectiveness of tax incentives to save”, Principal Investigator: Justin van de Ven, National Institute of Economic and Social Research.

1. Professor of Economics, University of Exeter and Research Fellow, Institute for Fiscal Studies.

From issue no. 158, July 2012, pp.13-14 and 24

Page Options