A macroeconomic approach to the income-tax work-effort relationship

Abstract

In this paper, we analyse the dynamic relationship between hours worked per employee (per self-employed) and marginal income tax-rate shocks in terms of both a comparative-dynamics model and a stochastic general equilibrium econometric model. The econometric model is estimated for Germany, UK and USA over the post-1960 period using the GMM estimation technique. Estimates in both models show that increases in the marginal income-tax rate exert negative effects on hours worked by both employees and the self-employed, but the response of the employees who are subject to tax withholding is stronger than the response of the self-employed.

JEL Classifications:

• E1
• E6

Keywords:

• hours worked per employee and self-employed
• tax withholding
• incentive to work
• labour supply
• marginal income tax rate

Notes

1. ^{For the sake of simplicity, the terms ‘work effort’ and ‘working time’ are used interchangeably in the present text to indicate weekly hours of work. In the real world, employees subject to contracts cannot vary working time, but they can vary effort, unless they are perfectly monitored.}

2. ^{As becomes evident from the above, our analysis will focus on the effects of an income tax on hours worked and no attempt will be made to examine the effects of the employee social security contributions. The underlying reasoning is that the response of working hours to increases in social security contributions is expected to be substantially milder than that of an income-tax increase, because of the redistributive or reciprocal nature of the former, as well as of the income-tax progressivity.}

3. ^{In Germany: the unemployment rate – to account for the global reductions of working time in the 1980s and 1990s to reduce unemployment – together with unionization, capital utilization, real unit labour costs – which accompanied the reduction in official weekly working hours in the period 1984–1994 – and a dummy variable to control for the effects of reunification. Efforts to econometrically estimate relationships between variables such as employment rates and real wages for Germany over a period covering both the pre- and the post-unification years is expected to meet serious problems, because of the different experiences of the two parts of this country. To cope with this problem, we ran two separate regressions for the equations of our model. The first set of GMM estimates used data solely for West Germany over the period 1960–1990; the second set was based on combined data for both West Germany up to 1990 and the reunified Germany over the period 1991–2007, with a dummy variable to capture the effects of reunification. Both sets of GMM estimates gave parameter values of the same sign and the same order of magnitude for the crucial variables of the model. Table 3 displays the estimates of the second set, which refers to the entire period 1960–2007. In the UK: the ratio of unionized workers to their total number – to account for the strength of labour unions – the net replacement ratio – to catch the disincentive effects on hours worked – and two dummy variables: the first for the period 1999–2003 (to account for the effects of the Working Time Directive on hours worked) and the second for the period 1980–2004 (to account for the impact of all the reforms made to improve work incentives). In particular, the second dummy variable was introduced to account for the fact that the UK launched several initiatives in the period 1980–2004 to increase labour market attachment of the unemployed. A series of employment laws reduced employees’, and especially unions’ bargaining power. Wage Councils were largely abolished, welfare benefits reduced and eligibility tightened. The most important of these initiatives, which cannot be captured by the tax variable, are the following.In the early 1980s, marginal withdrawal rates could exceed 100%, creating strong disincentives to work. This anomaly was tackled in 1988 by calculating entitlement on net rather than gross income.The November 1994 budget introduced changes to employers’ national insurance contributions (NICs) to favour employment of the part time, the low paid and the long-term unemployed. The same budget introduced: (i) nationwide extension of the ‘workwise’ and ‘1-2-1’ schemes; (ii) extension of the ‘Community Action’ scheme; (iii) extension of the ‘Work Trials’ scheme; (iv) nationwide availability of the ‘Jobfinders Grant’.In 1995, Family Credit and the Disability Working Allowance offered an extra UK£10 a week to claimants working for more than 30 hours a week.In 1995, Invalidity Benefit was replaced by Incapacity Benefit, which applied a tougher medical test to assess incapacity and eligibility for benefit.In 1996, the means-tested component of the Jobseekers’s Allowance (JSA) replaced ‘Income Support’ as a safety-net benefit with a marginal withdrawal rate of 100%. To counter the disincentive to work, a ‘back-to-work bonus’ was introduced.In the period 1998–2001, the UK initiated important active labour market programmes to reduce unemployment and inactivity. They are welfare-to-work programmes under the umbrella of the ‘New Deal’, which gives special attention to disadvantaged groups (young people, the long-term unemployed, lone parents, disabled people).The Working Families Tax Credit (WFTC) was introduced in 1999 to provide in-work financial support for families with children, in order to address the issue of workless households.In 2003, the government combined several parts of the tax and benefit system that supported families, including the WFTC, and replaced them with the Child Tax Credit and the Working Tax Credit. For a detailed discussion of the above measures, see OECD Economic Surveys, UK ( 1995 , 1998 , 2009 )}.^{In the US: employment growth, deviations of employment from trend and a lagged dependent variable – to capture cyclical influences – as well as a dummy variable for the period 1996–2003 to capture the effects of the 1996 Welfare Reform Bill and the 1998 reauthorization of Head Start by the Congress (because poor families with children are now better able to work since young children are at publicly funded schools, rather than privately funded daycare).}