The impact of social security contributions on earnings: Evidence from administrative data in France, Germany, the Netherlands and the UK

Surprisingly little is known about how social security (National Insurance) contributions (SSCs) affect earnings - for example, if employers' contributions are increased, to what extent and how quickly do they pass that on to employees in the form of lower earnings? At a time when governments in some countries are increasing SSCs in order to reduce budget deficits while others are moving away from employer SSCs amid concerns that they are detrimental to employment, having an understanding of their true effects is vital.
This project aims to provide new evidence on the effects of SSCs on earnings in France, Germany, the Netherlands and the UK. The research will exploit large administrative datasets in each country that span several decades and track the same individuals over time: these data have never before been used in cross-country analysis and offer the potential to advance our understanding of the impact of SSCs substantially.

In particular, we will address two related questions: the effects of SSCs on individuals' and firms' behaviour (labour supply and labour demand responses) and the economic incidence of SSCs (i.e. who bears the burden of them in the form of lower net income). We will try to distinguish between short-run and long-run effects, look for differences according to who is nominally liable for the tax (i.e. the employee or the employer) and how closely contributions are linked to future benefit entitlements, and explore the institutional features (such as wage bargaining processes, trade unions and minimum wages) that might affect these impacts. Looking at different countries and at different time periods will help to shed light on all of this, but we will also use several different methodologies and examine what we can learn from the results of the different approaches.

First, we will take a high-level look at the relationship over a long period in many countries between aggregate economy-wide measures of earnings and previous years' SSC levels.

Second, we will examine how far people in the UK, Germany, France and the Netherlands tend to be bunched around particular earnings levels where SSC rates change, and assess what we can learn from that about the extent to which earnings respond to the incentives created by SSC regimes.

Third, we will focus on reforms to SSCs in the four countries, looking at subsequent earnings growth of those affected by the reforms compared with earnings growth among similar but unaffected groups.
Fourth, we will create an economic model of individuals' job search activity and simulate the effects of SSC reforms.

Finally, we will divide the population in each of the four countries into groups based on age, sex, etc, and estimate the relationship between the average earnings of each group (in each country, in each year) and the average SSCs they faced.

Taken as a whole, this research programme will provide an exhaustive analysis of what experience across several countries over several decades can tell us about the impact of SSCs on earnings, resulting in a deeper understanding and a rich evidence base to inform future policymaking.

The international relevance of the issues we examine and the cross-country nature of this research mean that the impact of this project will be felt internationally.

It will benefit academics, as described in the Academic Beneficiaries section - and any further research that is motivated or facilitated by our work may bring wider benefits of its own. But this project also has significant potential benefits directly for policymakers in the UK and elsewhere. At a time when governments in some countries are increasing SSCs in order to reduce budget deficits while others are moving away from employer SSCs amid concerns that they are detrimental to employment, having an understanding of their true effects is vital. Yet surprisingly little is known about how SSCs affect earnings. This project will yield new evidence on the effects of SSCs on earnings in France, Germany, the Netherlands and the UK. In particular, it will shed light on two related questions: the effects of SSCs on individuals' and firms' behaviour (labour supply and labour demand responses) and the economic incidence of SSCs (i.e. who bears the burden of them in the form of lower net income). We will try to distinguish between short-run and long-run effects, look for differences according to who is nominally liable for the tax (i.e. the employee or the employer) and how closely contributions are linked to future benefit entitlements, and explore the institutional features (such as wage bargaining processes, trade unions and minimum wages) that might affect these impacts. Taken as a whole, the research programme will provide an exhaustive analysis of what experience across several countries over several decades can tell us about the impact of SSCs on earnings and what factors the impact depends on, resulting in a deeper understanding and a rich evidence base to inform future policymaking.

Those outside government concerned with good policymaking and the impact of SSCs will also share these benefits: this includes business groups, trades unions, NGOs, think tanks and international organisations such as the OECD. The media and the public in general may benefit as we address issues at the centre of public debate and bring new evidence to the table.

Private firms and individuals might even benefit directly from a better understanding of how future changes to SSCs might affect their costs and incomes respectively - the extent to which the immediate effects of such changes might be offset (or even accentuated) by adjustments to earnings - as well as benefiting from any improvement in policy design.

Part of the benefits of this project will arise as the project goes on, since results from each stage of the project will be published as the later stages proceed. But the bulk of the benefits will be felt towards and after the end of the project, as the full picture becomes available and the different parts can be integrated into a unified narrative, and as the bulk of the dissemination activity described in the Pathways to Impact section takes place.