Measuring the marginal efficiency cost of redistribution in the UK

This paper estimates the marginal efficiency cost of redistribution (MECR) associated with a demogrant and an in-work benefit for the UK since 1979, taking account of extensive as well as intensive labour supply responses. The principal methodological advance in the paper is its greater allowance for heterogeneity in the population than previous work. The rate of tax on consumption expenditure is allowed to vary across households and overall tax rates are allowed to vary across all individuals in all years, using a microsimulation methodology for the calculations. This disaggregation makes a substantial difference to the results. The central finding of the paper is that the MECR is much lower for the in-work benefit policy than the demogrant. The efficiency loss associated with a marginal in-work benefit has consistently been low (and occasionally negative): even at its current 25-year high, the policy would cost losers only £1.30 per pound that the gainers gained. By contrast, losers from a demogrant would currently lose £4.30 per pound that gainers received, higher than at other times over the last 15 years but still well short of the peak of £8.02 seen in 1981. Although precise estimates are highly sensitive to the overall levels of tax rates and elasticities, and also to the composition of the overall labour supply elasticity, the principal finding of a stark contrast between the two policies is robust. The paper also examines the effect of redistribution within family types. The inwork benefit policy looks even more favourable if paid to (and financed by) only singles; it looks less favourable if implemented only for childless couples. Increasing in-work benefits and/or cutting tax rates for lone parents have provided opportunities for Pareto-improving reforms to the tax and benefit system for most of the period since 1979.