Overview
This series follows Dur, Harms, and Non’s Dutch randomized information experiment on whether mistaken effective marginal tax rates affect work hours. It moves from the tax-and-transfer slope on the next €100, through baseline errors and the two-wave test, to asymmetric belief updating, feasible work, and a conditional distributional implication. The evidence supports a short-run hours effect among some overestimators, especially at lower incomes, but does not show that information replaces tax reform, removes constraints, or reduces inequality by a measured amount.
Learning outcomes
- The relevant incentive is the effective marginal return on the next €100, not a statutory rate or average tax wedge.
- Dutch respondents held large, directionally mixed misperceptions, with lower-income respondents overestimating more on average.
- In the two-month follow-up, overestimators corrected their beliefs and increased reported work hours by about 1.45 per week.
- Health, care, study, and workplace constraints weakened the path from corrected beliefs to additional hours.
- The paper suggests a possible inequality reduction under a wage assumption; it does not measure that outcome directly.
Keywords
Preview every chapter · Unlock full text, audio, and video from chapter 4
PlansFAQ
What is an effective marginal tax rate?
An effective marginal tax rate is the share of the next gross-income increment absorbed by taxes or reduced transfers. It is household- and income-specific, so it differs from an average burden.
How is an effective marginal tax rate different from the OECD tax wedge?
An effective marginal tax rate measures the disposable-income change from the next increment for a particular household, while the OECD tax wedge averages employee and employer labor taxes net of cash benefits at a defined wage and household type. The measures answer different questions.
What did the Dutch experiment find about tax information and work hours?
It found that a range-based information offer improved measured accuracy and increased reported hours by about 1.45 per week among initial overestimators in the two-month follow-up. The underestimator hours reduction was small and statistically imprecise.
Why did clearer tax information not change every worker’s hours?
Clearer information could not change hours when health, childcare, caregiving, study, or workplace conditions limited a feasible adjustment. Constrained overestimators also updated less, although that subgroup analysis was not preregistered and constraints were self-reported.
Can correcting tax misperceptions reduce income inequality?
It could in the study’s setting because hours gains were concentrated among lower-income respondents, but the implication assumes hourly wages do not fall by more than hours rise and was not directly measured.