VoxEU Column Labour Markets Macroeconomic policy

Who really pays social security contributions and labour taxes?

In spite of its policy relevance, academics and policymakers cannot agree on who bears the brunt of a tax on labour. This column uses meta-regression techniques to argue that economic institutions, the tax wedge definition, and the time horizon are crucial in determining who actually pays. Results based on 52 empirical papers suggest that in the long run, workers bear between two thirds of the tax burden in Continental and Anglo-Saxon economies, and nearly 90% in Nordic ones.

For almost two decades, a common policy recommendation to boost job creation from academic and international institutions has been to reduce social contributions. In order to compensate for revenue losses – taking into account that ageing will pose additional challenges to financing many social policies, especially in the fields of pensions and healthcare – this policy recommendation often includes an increase in other sources of revenues, such as consumption or environmental taxes.

Behind this broad consensus lies surprisingly wide-ranging economic rationales. For some, reducing labour taxes is a means to reduce labour costs, favouring labour demand (European Commission 1994, OECD1994). For others, tax cuts would increase home pay, thereby increasing labour supply (Prescott 2004, Coenen et al. 2008). And for emerging economies, especially in Latin America, some authors have focused on tax cuts’ positive effects on formality (Levy 2008, Pages 2010).

The choice between these alternatives hinges upon our views on who actually bears the tax burden. In the case of employer social contributions, they can be borne by firms (reducing their after-tax profits), they can be 'shifted backwards' to employees (reducing net wages of their workforce) or 'shifted forward' to consumers (increasing the price level of their products).

It’s an empirical question

Ultimately, this is an empirical question. The empirical literature is far from been unanimous. Hamermesh’s pioneering survey (1993) on the economic incidence of payroll taxes concluded that results covered the whole range, from full to null shifting. More recent references have not solved this indetermination. To quote a few, results vary from full shifting (Gruber 1997 for Chile, Alesina and Perotti 1997 for a sample of highly centralised and decentralised OECD countries), to no shifting at all (Heckman and Pages 2003 for a sample of OECD and Latin American economies, Kugler and Kugler 2003 for Colombia)1.

This lack of consensus about the effects of labour taxes and social contributions can be explained by the many political and socioeconomic factors that affect the tax-burden distribution:

  • Economic institutions;

In particular, the degree of centralisation and coordination of wage bargaining and the effectiveness of the public sector (see Calmfors and Driffill 1988, Alesina and Perotti 1997, Daveri and Tabellini 2000);

  • The type of pension scheme and, more importantly, the perception of the link between taxes and pensions;

If agents perceive a full linkage effect, social taxes become a deferred salary (see Gruber and Krueger 1990, Gruber 1994a and 1994b, Disney 2004):

  • The revenues included in the tax-wedge definition, since the bases of indirect taxes, personal income tax, employee and employer social contributions differ (see OECD, 1990 and 2007).
  • The time horizon.

This is important due to the presence of nominal rigidities, which imply that shifting tends to differ in both the short term and the long term (Hamermesh 1993).

Meta-analysis

This situation – a lack of consensus and competing economic hypothesis – is the optimal field for applying meta-analysis techniques, a still relatively recent survey technique in economics (see the seminal paper by Stanly and Jarrel 1989)2. In a recent paper (González-Páramo and Melguizo 2012), we perform a meta-analysis of a large and representative set of papers which estimate the impact of social-security taxes and labour taxation on wages and labour costs. We have assembled a database of 52 papers, covering most of the OECD countries and some Latin American economies (see Melguizo 2009 for a narrative survey). For each, we characterised the most salient qualitative features by means of dummy variables, ‘moderators’ in meta-analysis jargon.

Our empirical analysis is based on the 124 estimates of the elasticity of wages to taxes, taken from the 52 empirical papers. On average, in the core sample a 1% increase in taxation reduces wages by 0.66%. Therefore, employees would bear nearly two thirds of social security contributions, in line with the conventional wisdom on distributive incidence. However, the literature exhibits a very large degree of variability, ranging from papers who actually state that workers shift part of their taxes (0.91) to others where net wages are reduced in more than twice the employer tax burden (-2.54). As shown in Figure 1, differences seem evident when differentiating between economic institutions, tax-wedge composition and short- versus long-run results.

Figure 1. Distribution of wage elasticity to taxes (elasticity of net wages to taxation)

Source: González-Páramo and Melguizo 2012.

The meta-regression analysis confirms these results (see Table 1 below). The economic effects of social contributions are sensitive to both moderators representing basic economic institutions (which can be summarised in three 'models’: namely Anglo-Saxon, Continental-Mediterranean and Nordic) and the tax wedge definition – in particular, the inclusion of indirect taxes. Moreover, the impact of taxes on wages differs in the short as well as the long-term. In our preferred specification, the elasticity of wages to taxes is -0.70 in the default option, i.e. a non-Nordic economy in the long run. Therefore, workers bear 70% of taxes. In the Nordic economies the degree of shifting is close to full (-0.88), so all tax changes are almost entirely offset by a wage variation. The impact of taxes on wages is quite sensitive to the time horizon. The degree of shifting is much lower in the short run, and workers bear less than half of the tax burden. Finally, although not included in our preferred specification, consumption taxes may be more prone to shifting. The robustness checks controlling for quality of studies that were carried out – which used the complete set of results, alternative estimators and extreme-bounds analysis – confirmed these results.

Table 1. Meta-regression of the elasticity of net wages to taxes

Source: González-Páramo and Melguizo 2012.

Policy implications

The standard caveats about interpreting econometric results apply here, all the more so since a significant part of the differing conclusions found in the empirical literature remains unexplained. Bearing this in mind, some economic policy implications can be drawn from these results:

  • Policymakers should be aware that even if higher social-security contributions have a limited effect on employment in the long run, these are not entirely negligible, especially in the short run;

This empirical finding poses some limits to strengthening social protection systems via higher revenues.

  • Intermediate results point to significant employment gains from a revenue-neutral tax reform, i.e. increasing consumption taxes and lowering labour taxes.

Obviously, this tax-mix rebalancing might also involve some problematic aspects, namely a short-term inflationary effect and a change in income distribution, which should be evaluated.

  • Our estimates support the conclusion that taxation in Nordic economies, characterised by a high coordination of wage bargaining and effective public sectors, is more employment-friendly.
Conclusion

Returning to our initial question, ‘what would be the effect of reducing social security contributions and labour taxes?’, our answer is that it depends. Our paper shows that effects on labour would be stronger in Continental and Anglo-Saxon economies, while Nordic ones would see effects concentrated on net wages. Concerning the informality dimension – a pressing issue in the economic debate in Latin America – it seems that more evidence is needed.

References

Alesina, A and R Perotti (1997), “The Welfare State and competitiveness”, The American Economic Review 87(5), 921-939.

Calmfors, L and J Driffill (1988), “Centralization of wage bargaining”, Economic Policy 6, 13-61.

Coenen, G, P McAdam and R Straub (2008), “Tax reform and labour-market performance in the Eurozone. A simulation-based analysis using the New Area-Wide Model”, Journal of Economic Dynamics and Control, 32, 8, 2543-2583.

Daveri, F and G Tabellini (2000), “Unemployment and taxes. Do taxes affect the rate of unemployment?”, Economic Policy, 30, 48-104.

Disney, R (2004), “Are contributions to public pension programmes a tax on employment?”, Economic Policy, 19, 267-311.

European Commission (1994), Growth, competitiveness, and employment. The challenges and ways forward into the 21st century, Brussels. 

González-Páramo, J M and A Melguizo (2012), “Who bears labour taxes and social contributions? A meta-analysis approach”, Forthcoming in SERIEs.

Gruber, J (1994a), “The incidence of mandated maternity benefits”, The American Economic Review, 84, 3, 621-641.

Gruber J (1994b), "Payroll taxation, employer mandates and the labour market: theory, evidence and unanswered questions", mimeo.

Gruber, J (1997), “The incidence of payroll taxation: evidence from Chile”, Journal of Labor Economics, 15, 3, Part 2, S72-S101

Gruber J and A B Krueger (1990), “The incidence of mandated employer-provided insurance: lessons from workers’ compensation insurance”, NBER Working Paper 3557.

Hamermesh, D S (1993), Labor demand, Princeton, Princeton University Press.

Heckman, J J and Pagés, C (2003), “Law and employment: lessons from Latin America and the Caribbean. An introduction”, In J J Heckman and C Pagés (eds.), Law and employment: lessons from Latin America and the Caribbean, 1-107, Chicago, Chicago University Press.

Kugler, A and M Kugler (2003), “The labour market effects of payroll taxes in a middle-income country: evidence from Colombia”, CEPR Discussion Paper 4046.

Levy, S (2008), “Good intentions, bad outcomes. Social policy, informality and economic growth in Mexico”, Brookings Institution Press, Washington, DC.

Melguizo, A (2009), ”¿Quién soporta las cotizaciones sociales empresariales y la fiscalidad laboural? Una panorámica de la literatura empírica”, Hacienda Pública Española / Revista de Economía Pública 188-(1/2009), pp.125-182.

OECD (1990), “Employer versus employee taxation: the impact on employment”, OECD Employment Outlook, 157-177, Paris, OECD.

OECD (1994), The OECD Jobs Study, Evidence and explanations, Part II: The adjustment potential of labour market, Paris, OECD.

OECD (2007), “Financing social protection: the employment effect”, OECD Employment Outlook, 57-206, Paris, OECD.

Pagés, C (2010), The age of productivity. Transforming economies from the bottom up, Inter-American Development Bank and Palgrave Macmillan, New York, NY.

Prescott, E C (2004), “Why do Americans work so much more than Europeans?”, Federal Reserve Bank of Minneapolis Quarterly Review, 28(1), 2-13.

Stanley T D and S B Jarrel (1989), “Meta-regression analysis: a quantitative method of literature surveys”, Journal of Economic Surveys, 3(2), 161-170.


1 Latin American economies are particularly interesting due to the implementation of structural pension reforms (starting in 1981 in Chile, and followed in the mid-1990s in Argentina, Colombia, Mexico and Peru among others), which legally changed the nature of social security contributions from being a tax to a deferred salary.

2 In contrast to narrative surveys, meta-analysis revises relevant empirical literature in a more formal and objective way. First, it demands an exhaustive compilation sample of the literature which share the relevant dependent variable; in this case, the degree of social contributions backwards shifting (proxied by the estimated elasticity of net wages to taxation). Second, it involves a neutral approach to competing hypothesis, defining a general set of the ‘moderators’, that is, the quantitative and qualitative features in the different studies that could be influencing the impact of taxation on wages (tabulated, in practice, using dummy variables. And third, the meta-regression of the dependent variable on these moderators will show both the consensus result of the whole empirical literature after controlling for methodological differences, and which of the latter variables matter.

15,013 Reads