Designing Skill-Friendly Tax Policies
OECD Skills Strategy Spotlight No. 6
> Also available: français
March 2014 - The design of the tax system is a key driver for the skills outcomes of OECD economies and thus impact on employment, entrepreneurship and inclusive economic growth. Of course, the decision by individuals and companies to undertake education and training is influenced by more than just the tax code. Nonetheless, taxes have significant impacts on the finances of workers, students and companies. Everyone pays taxes that can rise or fall depending on their earnings before, during and after periods of education and training. Taxes affect the economic costs and returns to investments in skills and can therefore affect incentives to “upskill” or not. Moreover, taxes play a key role in influencing the decision to participate in the labour market or not, and thus to activate skills in the labour market. This Spotlight outlines some of the ways in which taxation can affect skills development decisions, as well as some key considerations in designing skill-friendly tax policies. We focus on skills policies targeted at third-level and adult education.
What are the key characteristics of skill-friendly tax policies? In general, the objective of tax policy is to finance government spending in the manner that interferes with skills decisions as little as possible in order to prevent tax-induced under- or over-investment in skills. This objective should be the main consideration when designing good tax policy. If, however, other features of the skill system are leading to inefficient or unfair skills outcomes, the tax code can also be used to impact skills investments of both employers and individuals, through different rates, deductions and credits. In addition, policymakers must consider the role of the tax code as it relates to other parts of the skills system such as the financing of education and skills formation. Other policy goals, such as the broader efficiency and fairness of the tax code, as well as the compliance and administrative burden and the tax revenue needed in a given year, must also be taken into consideration.
How can we know how the tax system is affecting skills? The key metric of a tax code’s effect on skills is the Effective Tax Rate on skills – how much, all things considered, does a tax code discourage or encourage people and businesses to invest in skills, compared to a world without tax? The literature in economics suggests that too-high taxes on physical capital can discourage investment. Human capital (skills) is no different in this regard. While taxation of physical capital has been an object of significant study for many years, policymakers are now also recognising the important ways in which taxes can affect investment in skills too.
How does the tax code affect skills development decisions?
The tax code can affect incentives to invest in education and training by influencing the costs and benefits of these investments. This can be the case for individuals through the income taxes and social security contributions they pay, and for companies through their corporate taxes and employer social security contributions.
How does the tax code affect individuals’ skills decisions?
For individuals, one of the main reasons to invest in skills is because these skills will be rewarded through higher incomes later. The tax code reduces this benefit as income taxes reduce the economic returns, and thus incentives to invest in skills and training. In addition, tax progressivity (where higher income levels are taxed at higher tax rates than lower income levels) means that those considering investments in skills – and expected higher incomes to follow – will face steadily higher tax rates. The more skilled workers become, the higher tax rate they pay. Thus tax progressivity can be an element of the tax code that reduces incentives to invest in better skills. Of course, progressive tax systems also help reduce earnings inequality and thus contribute to inclusive growth. The various effects of progressivity, including positive impacts on equality and potentially negative impacts on skills investments, must thus be balanced carefully by policymakers.
A second channel by which the tax code can reduce the benefits to education and training is by reducing skill activation. Taxes on labour can reduce workers’ incentives to work extra hours, return or stay longer in the labour market, and indeed, to work at all. Where taxes on labour – especially at lower ends of the income distribution – are too high, skills can remain inactive in the work force. Of course, this can have knock-on effects on skill development. If workers do not expect to use their skills much, they have less incentive to develop them. Thus too-high shares of labour taxes (compared to, say, goods and services taxes or property taxes) in the tax mix can be detrimental to skill formation.
Tax policies can also reduce the costs of skills investments both directly and indirectly. Indirectly, costs of education are reduced through the tax code’s effect on foregone earnings. For many of those considering investing in skills, especially in later life, a major cost is not the direct costs (tuition fees, textbooks and materials and so on) but rather the lost earnings involved in working less (or not working at all) while studying. Paying taxes, as mentioned, reduces the benefits of working. The income tax code affects these lost earnings in that those who are paying more in income tax will have greater incentives to move into education. This is especially so if the student has their own savings, or is financed by parents or is married to a partner who earns income to fund living expenses while studying. On the other hand, in many OECD economies, receipt of social benefits such as pensions or unemployment benefits is conditional on social security contribution history. Taking time out of work to invest in education and training can reduce access to these benefits, or reduce their value.
Directly, tax systems can reduce the costs of investing in skills. Educational spending can be seen as a ‘cost’ of earning income, and thus like other costs, be subject to tax relief. This can happen through tax credits and tax allowances (also known as tax deductions) that reduce the tax owed based on the amount spent on education. Tax Allowances are subtracted from the amount of income that is subject to tax. Tax Credits, in contrast to tax allowances, are subtracted from the amount of tax owed.
Targeting low-income households
A potential problem with some of these tax measures is that by reducing taxable income, they provide greatest benefits to those who have large tax bills in the first place. As these people can be those on higher incomes, these skills-friendly tax measures can be regressive – those who pay more in taxes receive more value from tax allowances than those who pay less. This can be avoided by using tax credits, which reduce tax liability and whose value does not depend on the level of taxable income, especially if they are designed as “refundable tax credits”. With these credits, if a taxpayers’ tax liability does not exhaust the value of the credit – because they do not owe enough in tax to exhaust it – they are refunded the remaining value of the credit. In this way, everyone receives the full value of the educational credit, regardless of their income. These can be particularly beneficial in reducing the costs of education for households with lower incomes who may have very little tax liability.
Distinguishing between education for employability versus consumption
A second potential problem with these kinds of credits is distinguishing between education that really is an ‘investment’ in productivity and education that is a form of ‘consumption’ as a form of leisure activity. To boost growth and productivity, there is a strong case for the tax code to exempt spending on skills investments. However, consumption education spending, like any other consumer good, should not be subject to tax relief. Distinguishing between these two kinds of education spending can present administrative challenges for governments. From a skills perspective, it is probably best to keep eligibility requirements for credits and deductions inclusive (i.e., going beyond traditional third-level education to include adult education), and risk subsidising some consumption at the margin, instead of having them be too stringent, and risk reducing skill development.
How does the tax code affect companies’ skills decisions?
For companies, the tax code can affect the decision to train workers through the company’s corporate tax liability. Corporate deductibility of training costs is widespread across the OECD. These involve the amount of corporate income that is taxable being reduced by the amount a firm spends on training. Most countries allow employers to make this reduction immediately – in the same year the money is spent on training. This is in contrast to physical capital, where these deductions take place over time as the physical asset loses its value (this is known as tax depreciation). This provides a relative advantage to human capital investment over physical capital in many tax systems.
The generosity of these deductibility provisions varies from partial deductibility (only some fraction of the training costs is deductible) to more than full deductibility. Some countries, such as Austria, provide still further incentives beyond full deductibility (e.g. 110% of training costs) to firms to encourage investments in skills. This can sometimes be a necessary response to factors that make human capital less attractive compared to physical capital. For example, poaching concerns – that newly-skilled employees may leave the firm after training, taking the value of their investment with them – may reduce the level of skill investment below the level which would be most efficient. Skill-friendly corporate tax provisions can be a response to some of these kinds of frictions and market imperfections.
Care must be taken, however, in designing these skills deductibility measures:
- To ensure that providing training does not become an untaxed form of compensation (i.e. a fringe benefit) for employees. Firms may subsidise or provide valuable training either in work or outside work as a substitute for wages, which might not be subject to income or payroll taxes. This is of special concern if the training is essentially a form of consumption for the worker, and not an investment. Authorities should therefore ensure that the kinds of skill development that the tax code is subsidising are really adding productivity for the firm and the worker, and are not just untaxed compensation.
- To reduce windfall gains – where tax reductions are subsidising firms who would have undertaken certain training even in the absence of a tax break. These firms could then receive an extra tax benefit, while the skills impact could be marginal. Such tax breaks cost the government revenue, which means higher taxes must be levied elsewhere.
- To target corporate tax deductibility towards those firms where the skills impact is largest. Small and young firms, which often generate many new jobs, may not have turned a profit yet or may have low profits. These firms may thus have little or no corporate tax liability. Thus - just as poorer families may benefit little from personal income tax reductions for skills - these firms may benefit little from corporate income tax reductions for skills too.
- To ensure that the tax break really produces additional training, not simply a change in the kind of training taking place. Where the tax code incentivises some kinds of training over others, or incentivises training some workers over others, firms may simply shift the kinds of training they provide, or shift the workers they train. But the overall level of training may not increase. To get value-for-money from credits and deductions, the focus must be on encouraging training that is not already taking place.
Another potential approach to incentivising firms to invest in skills is to reduce employers’ social security contributions (or payroll taxes) instead of reducing corporate taxes. This can be done for firms that train their workers, or for firms that hire apprentices, young people or long-term unemployed workers. Such tax breaks have the benefit that they will affect firms in proportion to their labour costs, not in proportion to their profitability. A drawback to these provisions, however, can be that the funding situation of the social security system can be undermined.
All these policies need to be regularly evaluated to determine which kinds of tax provisions are useful in encouraging firms to invest in skills. They should also be compared to, and seen in the context of, other policies such as grants and loans to firms, or indeed direct state provision of education. Sunset clauses, by which polices have to be explicitly continued by future governments, could be a means to ensure that that these evaluations take place.
OECD is a recognised source of comparative skills data and policy analysis and is working to develop methodologies to better analyse skills demand and supply of countries, including in the area of tax policy.. The OECD Skills Strategy sets out a coherent framework for developing national and local skills strategies based on three pillars: developing relevant skills, activating the supply of skills and the effective use of skills. The OECD Survey of Adult Skills (PIAAC) provides unique comparative data on the skills of adults in literacy, numeracy and problem solving in technology-rich environments.
For further reading
- Brys B. and C. Torres (2013) Effective Personal Tax Rates on Marginal Skills Investment in OECD Countries, OECD Taxation Working Paper 2013/13, OECD Publishing, Paris.
- OECD (2013), Taxing Wages 2013, OECD Publishing, Paris.
- OECD (2011), Taxation and Employment, OECD Tax Policy Studies No. 21, OECD Publishing, Paris. Torres, C. (2013), Taxes and Investment in Skills. OECD Taxation Working Paper 2013/13, OECD Publishing, Paris.
This paper is published under the responsibility of the Secretary-General of the OECD. The opinions expressed and arguments employed herein do not necessarily reflect the official views of OECD member countries.