News

Working Paper No. 1/2026 “Wage dynamics and the role of the personal income tax”

21 July 2026 | This Working Paper examines how the earnings of private-sector employees in Italy have evolved since the 1990s, and the role played by reforms of the personal income tax (IRPEF) in supporting net earnings and curbing inequality.

Italy’s earnings growth has been particularly weak by international standards. According to OECD data, real gross earnings per full-time-equivalent employee fell by 1.6 per cent between 1990 and 2024, whereas they rose significantly in the other main advanced economies. The spread of part-time work compounded this weak wage growth, further reducing annual earnings and widening their dispersion. Against this backdrop, the IRPEF reforms introduced since the 2000s – and especially from 2014 onwards – have gradually strengthened support for lower income employees, enhancing the tax’s redistributive role for these taxpayers.

The analysis draws on INPS’s LoSaI longitudinal sample, which contains detailed information on the working careers and earnings of private-sector employees over the 1990-2018 period. The final dataset comprises around 4.8 million observations for more than 460,000 non-managerial workers. In order to extend the analysis to the 2019-2026period, earnings were updated using the collectively bargained pay increases by sector and occupational category, with the composition of the population held unchanged, so as to isolate the effects of the most recent tax reforms.

The data confirm the decline in real gross earnings in the private sector: average earnings fell by 6.6 per cent between 1990 and 2026. The contraction was concentrated in the lower part of the distribution, where the 10th percentile fell by 40.8 per cent, while the 90th percentile rose by 3.3 per cent. After 2021, the acceleration of inflation caused a further loss of purchasing power across the whole earnings distribution. The data also reveal growing segmentation of the Italian labour market along three dimensions: sector of activity, type of contract and occupational category. At the sectoral level, between 1990 and 2026 real gross earnings rose by 16.5 per cent in industry and construction, while they fell by 20.9 per cent in services. Wide gaps also persist between full-time and part-time workers, and between white-collar and blue-collar workers, with the latter recording the least favourable earnings trends. Overall these factors have widened wage inequality over the considered period.

Over the same period, the IRPEF on employment income was profoundly reshaped. The reforms progressively reduced the burden on low and middle incomes, especially after the introduction of the IRPEF bonus in 2014 and the 2025 reform, which incorporated the social security contribution relief into the structure of the tax. Conversely, for higher incomes the effective tax burden rose, mainly as a consequence of fiscal drag: income brackets and the other monetary parameters of the tax had not been indexed to inflation. The threshold separating the two effects is around €35,100 per year at 2026 prices. Below this level, the 2026 system is more favourable than that of 1990, above it is more burdensome (Fig. 1). Over the same period, the share of workers in the lowest income bands also rose, mainly as a result of the spread of part-time work.

 

Changes to IRPEF feed through into net employment incomes. The gaps between different groups of workers narrow in the move from gross to net incomes, compressing the dispersion. Nevertheless, while it offsets some gaps – such as those between full-time blue-collar and white-collar workers – the tax does not eliminate the inequalities generated by the labour market.

The rise in the IRPEF’s redistributive effect also emerges from summary inequality indicators. Between 1990 and 2026, the Gini coefficient of gross private-sector employment income rose from 0.330 to 0.403, while that of net income grew more moderately, from 0.302 to 0.338. As a result, the redistribution index – given by the difference between the two – doubled, rising from 0.028 to 0.065 (Fig. 2).

 

To understand which were the drivers of the increase in the IRPEF’s redistributive capacity over the 1990-2026 period, four factors have been identified: (i) tax reforms targeting low and middle incomes (at or below the €35,100 threshold at 2026 prices); (ii) tax reforms affecting higher incomes (above the threshold); (iii) fiscal drag, caused by the lack of indexation of the tax’s parameters; and (iv) changes in the income distribution and in the composition of the workforce that are not attributable to fiscal drag, linked mainly to the spread of part-time work. These effects are estimated with counterfactual simulations, carried out through a tax-microsimulation exercise, which isolates the contribution of each component while holding the others constant. This makes it possible to separate the impact of legislative changes from that of inflation and of labour-market change.

As is well established in the literature, the redistribution caused by a tax depends on its progressivity (measured by the Kakwani index) and on the level of taxation (measured by a tax-level index, which is a function of the average tax rate). For a given degree of progressivity, a tax with a higher average rate redistributes more; for a given average rate, a more progressive tax redistributes more. The effect of each factor on the change in redistributive capacity can therefore be split into two components, one relating to progressivity and one to the average tax rate.

This decomposition is repeated for every pair of years in the period (1990-91, 1991-92, and so on up to 2026). The adopted structure ensures that the effects are additive over time, so that the entire change in redistributive capacity over the 1990-2026 period can be broken down into the sum of effects of each factor on progressivity and on the average tax rate for each pair of years.

The results show that the rise in redistribution over the 1990-2026 period is due mainly to the reforms targeting low and middle incomes, which account for 115.8 per cent of the overall increase. The changes affecting higher incomes, by contrast, reduced redistributive capacity by 36.7 per cent, while fiscal drag and shifts in the income distribution made positive contributions, of 7.4 and 13.5 per cent respectively. Overall, tax reforms explain around 80 per cent of the increase in redistributive capacity observed over the period (Fig. 3).

 

The rise in redistribution was driven above all by the tax becoming more progressive. The Kakwani progressivity index rose from 0.131 to 0.396 and, on its own, accounts for around 152.4 per cent of the overall change in the redistributive effect. This increase was partly offset by the fall in the average tax rate – a result of the reliefs introduced over time – which dampened the growth of redistribution while supporting workers’ disposable income. The reforms targeting low and middle incomes worked mainly by making the IRPEF more progressive and lowering the taxburden on these taxpayers. The measures aimed at higher incomes, by contrast, reduced both progressivity and the average tax rate, producing on balance a negative contribution to redistributive capacity. Fiscal drag pulled the two components in opposite directions: it raised the average tax rate but reduced progressivity, with a net effect that was positive but limited. Labour-market changes also contributed to increase the redistributive capacity of the system. The spread of part-time work, and the resulting rise in the share of workers in the lowest income bands, widened the pool of beneficiaries of the tax-support measures. Indeed, in a progressive system a larger share of low-income taxpayers produces mechanically – at unchanged rules – an increase in the tax’s redistributive effect.

The decomposition in annual changes allows the effects of the various factors to be identified over time and their contributions to be aggregated by legislative term. The increase in redistributive capacity is concentrated above all in the 17th legislative term (2014-18) – when the IRPEF bonus was introduced – which accounts for around 39 per cent of the overall change, and in the 19th legislative term (2023-26), to which a further 26 per cent can be attributed, thanks to measures targeting low and middle incomes – among them the transposition of the social security contribution relief into the IRPEF. A smaller contribution (14 per cent) related to the 14th legislative term (2002-06).

The analysis therefore shows that the IRPEF has played a growing role in offsetting the dispersion of gross employment incomes, but it cannot establish a causal link between wage dynamics and tax reforms. From a theoretical standpoint, the existence of circular effects cannot be ruled out: the availability of fiscal tools to support lower incomes may have shaped wage dynamics through indirect channels, for instance by easing bargaining pressure. Accommodative fiscal policies and wage moderation may thus have worked together as parts of an implicit equilibrium among the actors in the labour market.

The results also point to some weaknesses in the current redistributive model. Progressivity was strengthened mainly by reducing the tax burden on low and middle incomes and concentrating it on upper-middle incomes, which may weaken incentives for skilled work and leave the system more exposed to fiscal drag. The differences in treatment between employees and other taxpayers, such as the self-employed and pensioners, have also widened. The IRPEF’s evolution is thus an ex post response to the growing dispersion of incomes, but it does not address the structural causes of wage stagnation and of the increasing precariousness of employment.

These considerations suggest that the current redistributive model may have reached its limits. Recent developments point in the same direction. On one side, fiscal drag – so far neglected by policy interventions – could re-emerge with renewed intensity in the inflationary environment triggered by the new geopolitical crisis. On the other, the use of substitute tax regimes for the wage increases arising from collective bargaining renewals, introduced by the 2026 Budget Law, reveals an inconsistency in the system: taxation on wage increases is significantly higher than that on income already earned, to the extent that these increases were removed from the ordinary tax regime to avoid cancelling out the effects of the renewals. Further increase in redistribution would probably require complementary measures: reviewing the preferential regimes that erode the ordinary system’s tax base, broadening the tax base by tackling tax evasion, and structural policies to counter wage stagnation and the rising precariousness of employment – possibly alongside income-support measures operating on the expenditure side.