Australia, like many countries, wrestles with a familiar trade-off in tax design: simpler systems cut compliance costs and widen access, but they may also open doors to misreporting.
Our new research studies this trade-off in France’s self-employment tax system and quantifies how much people value simplicity—and how much of the demand for “keeping it simple” is really about making evasion easier. We believe the lessons travel well beyond France.
A natural experiment in “simple” taxes
From 2006 to 2015, France offered self-employed workers three personal tax regimes that differed starkly in complexity.
The standard regime required full accounts on self-employed revenues and costs, with taxation integrated with other income and tighter verification. The simplified regime taxed a fixed share of revenue under a progressive schedule with lighter reporting. The super-simplified regime, introduced in 2009, replaced income tax and social contributions with a single low flat tax on gross revenue and the lightest reporting of all.
Eligibility for the two simpler regimes depended on staying under a revenue threshold that varied by activity and year. These thresholds create what economists call notches: cross the threshold and administrative burden rises, average liabilities may increase, and scope for misreporting falls; sit just below it and life is easier. This design lets us observe how keen people are to stay in simpler regimes, and why.
Using the universe of French income tax returns from 2006 to 2015, we follow self-employed individuals year by year and examine behaviour around eligibility thresholds. We combine classic bunching methods with dynamic evidence and a structural model to disentangle three forces: real changes in revenue, a preference for simplicity (lower hassle), and misreporting when it becomes easier to do so.
How taxpayers actually respond
We observe sharp spikes in the number of self-employed just below the eligibility thresholds for both simplified regimes, with a much larger spike for the super-simplified regime. This holds across activities and over time, even when pure tax rate differences at the threshold are small. In short, taxpayers strongly prefer simpler regimes, and the preference rises with simplicity.
Is this driven by misreporting? Our longitudinal evidence points to manipulation of reported amounts to the tax administration rather than changes in actual revenues from economic activity. Individuals who locate just below the threshold in a given year deliberately slow the growth of their reported self-employment income to remain under the threshold in the next year—a pattern absent among comparable taxpayers in the standard regime.
We also find additional fingerprints of manipulation near thresholds. Amounts ending in zero and round multiples of €100 are markedly more frequent, especially under the super-simplified regime. In couples where both partners are self-employed, the low earner’s reported revenue rises as the high earner approaches the ceiling, consistent with shifting income across partners to preserve eligibility. Reported wage income also increases alongside self-employment income at the threshold, suggesting some pay is relabelled as wages when eligibility for the simple regime is at risk.
Taken together, these patterns indicate that simplicity also makes evasion easier.
How much is simplicity worth—and how big is evasion?
Our structural estimates separate real responses, hassle costs, and misreporting. Real production responses at the margin are negligible: people aren’t changing what they produce much; they are changing what they report.
The evasion elasticity is sizable, implying substantial under-reporting near ceilings—roughly €800–€1,200 in the simplified regime and €1,400–€1,800 in the super-simplified regime.
Taxpayers also exhibit a meaningful preference for simplicity. Converted to money, the average administrative-hassle saving from being in a simple regime is about €49–€70 per year (simplified) and €342–€495 per year (super-simplified) per self-employed worker—roughly 5–8 hours and 36–53 hours of work at 2012 French minimum wage level.
Why this matters for policy
The headline is not that simplicity is bad. It is that simplicity carries two values at once: it reduces administrative burden for taxpayers but also lowers the cost of evasion. Both forces are quantitatively important, and the evasion channel grows stronger as regimes become very simple. That is a design challenge, not an indictment.
Our estimates focus on behaviour at eligibility ceilings and on reporting choices; they do not quantify extensive-margin effects such as additional entry into self-employment or the formalisation of previously informal activity. Simple regimes may draw new entrepreneurs into the tax net, expand downstream GST/VAT and social-contribution bases, free up time for productive tasks, and even lift output through better matching, innovation, and customer-facing growth. These benefits lie partly outside our measurement window. In any cost–benefit calculus, they should be weighed alongside the evasion risks we document—arguing for enforcement tools that preserve the participation gains of simplicity rather than blunt them.
What follows from this in practice? Policymakers should expect bunching at ceilings and plan for it. Enforcement should target the evasion margin rather than the business itself: the evidence points to reporting responses—digit patterns, round numbers, and intra-household shifting—rather than large real output effects. Light-touch, data-driven checks for obvious manipulation near thresholds can deliver high bang-for-buck without undermining genuine simplicity.
Because micro-businesses are potentially family enterprises, eligibility rules designed at the individual level should account for spousal income-shifting, with simple, low-intrusion household-level cross-checks. Authorities should also watch for misclassification—“hidden employment”—by focusing narrowly on relabelling risks where incentives spike. When redesigning regimes, it is often better to avoid unnecessary notches: gradual phase-outs or multi-year averaging around eligibility can prevent abrupt jumps in reporting stringency and reduce the payoff to staying just below the line.
Bottom line
Simplicity is key—people value it. But the simpler you make reporting, the simpler you also make evasion, unless you design with the evasion margin in mind.
Our results from France suggest a balanced recipe: keep the administrative relief taxpayers demonstrably value; deploy surgical, data-driven checks around thresholds; address household shifting and misclassification with narrow, well-timed controls; and smooth eligibility to blunt the cliff. That way, policymakers can bank most of the gains from simplicity without paying too high a price in lost integrity.
Recent Comments