The 50p tax rate

Episode Summary

Executive Summary: The episode debunks headline-driven claims about tax, wages, child costs, and sickness benefits by showing how changing behavior, selective comparisons, and weak assumptions distort statistics. It explains why the 50p tax-rate debate is uncertain, why living-standards claims can both be partly true depending on framing, and why an expensive-childhood estimate is more marketing than public fact.

Main Topics: The 50p top tax rate and the 'moveable cake' problem (Priority: 5/5): The programme explains why comparing tax revenue before and after a rate change is not straightforward: taxpayers alter behavior in response to tax changes, so revenue is not fixed and can shift through forestalling and other responses. Ed Balls' claim about £10 billion extra tax (Priority: 5/5): It examines Labour's argument that updated HMRC data show the 50p rate raised far more than originally thought, and concludes the £10 billion figure was already in the analysis used to estimate the policy's impact. Living standards and wage statistics (Priority: 4/5): The show contrasts government and Labour claims about rising or falling living standards, showing how different inflation measures, time periods, and after-tax versus before-tax earnings can produce opposite narratives. The real cost of raising children (Priority: 4/5): It critiques a press-release-driven estimate that raising a child costs £225,000, arguing the figure relies on a worst-case scenario with many assumptions rather than an official average cost. Sickness and disability benefit headlines (Priority: 3/5): The episode challenges claims that new figures show millions of sickness benefit applicants being found fit for work, noting that overall claimant numbers changed little despite tougher assessments. Using statistics to plan a wedding guest list (Priority: 2/5): A lighter segment shows a couple using statistical modeling to estimate wedding attendance, illustrating both the value and the limits of prediction when real-world behavior deviates from assumptions.

Key Arguments: Tax revenue cannot be estimated by simply rerunning old tax rates on new income data because taxable income changes when tax rates change. Forestalling before the 50p rate and reverse forestalling before the 45p cut heavily distort short-run revenue comparisons. The HMRC estimate that the 50p cut would cost only £100 million already used updated self-assessment data, so Ed Balls's £10 billion discrepancy claim is mistaken. The true revenue-maximizing top tax rate is highly uncertain; the IFS says it could be anywhere from below 30p to above 75p. Government and opposition living-standards claims can both be selectively true depending on inflation measure, time window, and whether wages are measured pre- or post-tax. The broader household-income picture is worse than official wage claims suggest: median household income remains below pre-crisis levels. The child-cost figure is designed to sound alarming and depends on a very specific, high-cost scenario rather than an average family experience. The sickness-benefit statistics show little overall change in claimant numbers despite stricter tests, so dramatic 'crackdown' headlines overstate the effect. Statistical modeling can help with real-world planning, but unexpected behavior and unmodeled factors can still change outcomes.

Data Points: Tax revenue change under 50p rate: Almost £10 billion more - Claim made by Labour about taxes paid by those earning over £150,000 during the 50p rate years. Estimated direct cost of cutting 50p to 45p: £100 million a year - HMRC estimate used by George Osborne to justify the tax cut. Potential revenue loss if nobody changed behavior: £3.5 billion - HMRC's principal estimate of the gross revenue effect of the rate cut before behavioral responses. Forestalling amount for 50p introduction: £16–18 billion - HMRC estimate of income brought forward to avoid the higher 50p rate. Top-rate revenue-maximizing band: Below 30p to above 75p - Institute for Fiscal Studies estimate of uncertainty around the rate that maximizes revenue. Take-home pay growth: For many people, a little bit faster than inflation - Government claim about recent living standards. Labour claim on real wages: Over £1,600 a year lower - Labour statement about real wages under the government. Median household income since crisis: More than 6% lower - Institute for Fiscal Studies measure of household income after inflation. Child-raising headline cost: £225,000 - Liverpool Victoria/Cebr estimate used in the discussion. Childcare cost estimate: £66,000 - CEBR estimate from age two to five for 50 hours a week, 50 weeks a year. Education cost estimate: £73,000 - CEBR estimate by age 21, excluding private school; includes university-related assumptions. Private school extra cost: £120,000 - Additional cost if private schooling is included. Education-related extras: £43,000 - Includes uniforms, packed lunches, after-school clubs, private tuition, and other expenses. Typical annual child cost: Around £8,000 a year - Broad estimate cited after excluding the most expensive periods like childcare and university. Child-cost inflation rate: 2.1% last year - Reported increase in the estimated cost of raising children. Consumer price inflation: 2% - Used to show the child-cost estimate is only slightly above general inflation. Sickness/disability claimants: 2.6 million in 2008 to 2.5 million in 2013 - Overall claimant numbers changed little despite the new assessment system. Share found fit for work: 80% - Proportion of claimants passing the new sickness benefits test. Wedding invitations sent: 139 - Number of invitations posted in the wedding planning example. Wedding venue capacity: 110 people - Maximum comfortable capacity for the venue. Wedding attendance: 105 people - Final number of guests who showed up at the wedding. Original spreadsheet estimate for wedding: 97 people - Damjan and Joan's model-predicted attendance.

Pivotal Quotes: "What you're suggesting would work perfectly well if all the taxable income in the country was a sort of delicious, gigantic cake" — Tim Harford: Explaining why simple revenue comparisons fail when taxpayers change behavior. "The big issue here is the way that taxable income changes in response to tax rates." — Ruth Alexander: Core explanation for why tax-policy revenue estimates are uncertain. "The direct cost is only £100 million a year." — George Osborne (quoted in transcript): Justification for reducing the top tax rate from 50p to 45p.

Implications: Listeners should be skeptical of dramatic statistics, especially when they depend on selective baselines, hidden assumptions, or behavior changes. Policy debates on taxes, wages, and welfare need broader context than headline numbers provide.