Helen Miller (Institute for Fiscal Studies) considers when and how policy makers should use taxes to steer the industrial structure.
In the UK’s 256 page industrial strategy (Industrial strategy: building a Britain fit for the future), the words ‘tax’ or ‘taxation’ crop up only a handful of times. On the face of it, that’s perhaps not too surprising: tax is just one policy lever among many others, ranging from education, through infrastructure and regulation, to direct government investments in science. It’s understandable that the industrial strategy gives a nod to the role of an internationally competitive corporate tax rate in attracting multinationals and reinforces support for R&D tax credits but builds a strategic vision from policies that more directly steer the economy down a certain industrial path.
Yet the list of ways in which our tax system affects the UK’s industrial structure is long. Capital allowances favour investment in plant and machinery over investment in buildings and purchased intangible assets. The tax deductibility of interest payments favours debt over equity financed projects, while a set of venture capital schemes favour those new companies that can secure equity investments. A suite of policies favour ‘small business’ – where small is defined on turnover for the VAT threshold, on property rental value for business rates relief, on the size of shareholdings for entrepreneurs’ relief and on turnover and staff for R&D tax credits. And these are just the obvious ones.
Keep digging and you find a hefty tax penalty on employment relative to self-employment, an incentive to hang on to assets with capital gains until death, various subsidies to agriculture and to investments in the creative industries, VAT zero-rates that give some products a competitive advantage, and so on. By favouring certain types of business, asset, financing or ownership, we’re effectively favouring certain industrial structures, even if it is hard to look at a list like this and easily discern which industries are getting a relative leg up.
Tax might not have a starring role in the industrial strategy, but it affects our industrial reality. What’s more, the effects are usually unintended and often problematic. In light of the UK’s long tail of unproductive and poorly managed firms, are we sure that policies encouraging people to start and hang on to business should be celebrated? If we are serious about financial stability, is it right to overlook the tax induced debt bias we have created?
We shouldn’t expect to see detailed tax policies filling the glossy pages of an industrial strategy, but neither should we ignore the unintended effects that taxes can have on our industrial structure. We could start by giving more consideration to these effects when we design tax policy.
If we were designing taxes from scratch, a good rule of thumb would be to tax all profits (as opposed to turnovers) in the same way regardless of how they arose (i.e. no special rates for anything). This in turn would require deducting the costs of doing business – such as investing in assets – without discriminating between assets, sources of finance or ownership. Such a system would ensure that decisions were driven by what makes commercial sense rather than by the tax system. We know that taxes affect decisions – you can’t prevent all the hissing when plucking geese – but we at least want to minimise the extent to which taxes distort choices unless there are good reasons to think that varying taxes across activities can improve outcomes.
A neutral tax system has much to recommend it but there are reasons to deviate from neutrality by favouring some activities over others. Deviations can be justified when the benefits of a simplification (such as removing many firms from VAT) outweigh the costs of added distortions or as a mechanism to meet redistributive goals. Some distortions arise from a desire to be internationally competitive and may be a price worth paying to ensure that the UK is an attractive location for multinational investments. But, too often, the ways in which our tax system influences industrial structure arise from an ill-thought-out desire to boost activities that sound like the kinds of things we must want more of.
A more structured way to design tax policies that will influence the industrial structure – and to evaluate the long list of tax breaks and special allowances – is to ask the following three questions.
Markets aren’t perfect or completely free. Imperfect competition is the norm and government intervention already shapes how markets operate. But while the many ways in which markets fail to produce the best possible outcomes lead to multiple arguments for deviating from a neutral tax system, doing so in practice is fraught with the dangers of complexity, special pleading and unintended consequences.
The first step when contemplating using policy to manipulate an outcome is to specify precisely what it is about current market outcomes that’s undesirable. What do we want more or less of, and why? Wanting more robots or small businesses or UK-made films because they’re ‘good’ in some way isn’t a sound enough rationale. If they’re good, why won’t the current market create the number of robots or films that balance the overall costs and benefits to society? If people aren’t willing to start a business without a subsidy, what makes us think that such a subsidiary should exist?
One specific reason that markets may fail to produce the socially optimal outcome is ‘spillovers’ (also called externalities). When individuals or companies make decisions, they don’t take account of the costs or benefits that their actions have on others; for example, they don’t value the lessons that others learn when a new technology is researched or a new business concept tried. Spillovers can mean that the market will produce too much of some things (like pollution) and not enough of others (like R&D) and this can be the basis for possible government intervention.
Knowing that there is an outcome we’d like to change is only the first step. There are rationales for the government to try, for example, to reduce pollution, problem drinking and childhood obesity while boosting R&D and entrepreneurship. Tax may not be the best tool, however. The government can also ban or regulate certain activities (such as advertising junk food to children), adjudicate competition (for example, between established firms and start-ups) and make direct investments (such as in science facilities or lending to new businesses).
Broadly, taxes are a good tool when changing prices (such as making the creation of pollution more expensive or R&D investments cheaper) or rates of return (such as making the post-tax rate of return higher if income is taken in capital gains than in earned income), and lead to more (or less) of the activity that the market is currently under (or over) producing. For example, R&D tax credits reduce the price of investing in research and thereby give companies an incentive to do more. If we want very specific outcomes (such as investment in malaria drugs), more direct intervention (for example, funding research or providing prizes) can be better targeted. And since we usually operate tax and direct measures simultaneously (for example, the UK spends around £3bn on R&D tax credits and about the same on direct support for business R&D), we should consider how they interact and what purpose different policy levers serve.
There may be a clear rationale for a policy and a reason why tax is, in principle, a good tool to use. But we still need to go to the effort of designing a policy that works and that doesn’t cause more problems than it solves. Consider entrepreneurship. One can justify considering policies that reduce the cost of trying a new business venture; however, that absolutely does not mean that a set of large, across the board tax breaks to all small business owners is a well targeted policy. It certainly might help to boost some of the activities that have externalities, but can also create a whole bunch of other unintended side effects (such as tax motivated incorporation) along the way.
In policy discussion, there’s a disappointing tendency to overestimate the ability to fine tune outcomes with tax and dramatically underestimate the set of unintended ways that people respond to tax incentives. Also, having identified that the market may underproduce entrepreneurs, it doesn’t mean that more is always good – we can have too many people starting new ventures. Finally, if the actual concern about start-ups relates to access to credit, then we should address that problem directly – offering juicy tax breaks for activities that generate capital gains doesn’t help a business that can’t get a loan today.
There are serious costs of sloppy thinking at the design stage. Poorly designed policies lead to inefficiencies (which ultimately make us poorer), unfairness (for example, when two very similar people get very different tax rates) and complexities. What’s more, since poorly designed policies often don’t work, we end up compounding distortions by layering new policies on top without considering how various policies interact.
One way to guard against a proliferation of poorly targeted policies is to evaluate what’s working. R&D tax credits offer a glimmer of hope here – there’s been a fair amount of research on their effects and there’s evidence that they lead to more R&D. This doesn’t mean the policy is perfect – plenty of the cost goes to investments that would have happened anyway and in an international context we should consider competition with other countries. But at least we have an evidence base. Most policies have never been evaluated. In many cases, there is no satisfactory way to do an evaluation because no data was collected and policies were introduced in ways that don’t allow their direct effects to be teased out.
While building industrial strategy into tax design might sound quite exciting, we shouldn’t get carried away. The bar for deviations from a neutral tax system (i.e. one that doesn’t affect our industrial structure) should be high. This is not because a neutral tax system is perfect but because attempts to use tax to help some groups too often create more problems than they are worth.
When we find problems or outcomes we’d like to change, we should first look at current policy. Removing current distortions is a seriously underrated policy option but is often much more sensible than layering another distortion on top. For example, there has been much concern about the effect of business rates on the high street, with proposed policy solutions including reliefs for pubs, retailers and small business and new taxes on online sellers. Instead, we could just fix business rates by moving the tax away from business buildings to land values, thereby removing all distortions to competition. This approach would not only make the system simpler rather than more complicated but it would actually work – whereas the quick fixes may offer some short-term relief but leave the structural issues in place.
If we were more willing to remove policies that weren’t working, we’d not only get a fairer, simpler and more efficient tax system but we could be more confident in experimenting with new policy options. If every new tax break or relief is destined to be a permanent fixture, we need to be much more mindful of how it will help to shape the long-term structure of the economy.
Helen Miller (Institute for Fiscal Studies) considers when and how policy makers should use taxes to steer the industrial structure.
In the UK’s 256 page industrial strategy (Industrial strategy: building a Britain fit for the future), the words ‘tax’ or ‘taxation’ crop up only a handful of times. On the face of it, that’s perhaps not too surprising: tax is just one policy lever among many others, ranging from education, through infrastructure and regulation, to direct government investments in science. It’s understandable that the industrial strategy gives a nod to the role of an internationally competitive corporate tax rate in attracting multinationals and reinforces support for R&D tax credits but builds a strategic vision from policies that more directly steer the economy down a certain industrial path.
Yet the list of ways in which our tax system affects the UK’s industrial structure is long. Capital allowances favour investment in plant and machinery over investment in buildings and purchased intangible assets. The tax deductibility of interest payments favours debt over equity financed projects, while a set of venture capital schemes favour those new companies that can secure equity investments. A suite of policies favour ‘small business’ – where small is defined on turnover for the VAT threshold, on property rental value for business rates relief, on the size of shareholdings for entrepreneurs’ relief and on turnover and staff for R&D tax credits. And these are just the obvious ones.
Keep digging and you find a hefty tax penalty on employment relative to self-employment, an incentive to hang on to assets with capital gains until death, various subsidies to agriculture and to investments in the creative industries, VAT zero-rates that give some products a competitive advantage, and so on. By favouring certain types of business, asset, financing or ownership, we’re effectively favouring certain industrial structures, even if it is hard to look at a list like this and easily discern which industries are getting a relative leg up.
Tax might not have a starring role in the industrial strategy, but it affects our industrial reality. What’s more, the effects are usually unintended and often problematic. In light of the UK’s long tail of unproductive and poorly managed firms, are we sure that policies encouraging people to start and hang on to business should be celebrated? If we are serious about financial stability, is it right to overlook the tax induced debt bias we have created?
We shouldn’t expect to see detailed tax policies filling the glossy pages of an industrial strategy, but neither should we ignore the unintended effects that taxes can have on our industrial structure. We could start by giving more consideration to these effects when we design tax policy.
If we were designing taxes from scratch, a good rule of thumb would be to tax all profits (as opposed to turnovers) in the same way regardless of how they arose (i.e. no special rates for anything). This in turn would require deducting the costs of doing business – such as investing in assets – without discriminating between assets, sources of finance or ownership. Such a system would ensure that decisions were driven by what makes commercial sense rather than by the tax system. We know that taxes affect decisions – you can’t prevent all the hissing when plucking geese – but we at least want to minimise the extent to which taxes distort choices unless there are good reasons to think that varying taxes across activities can improve outcomes.
A neutral tax system has much to recommend it but there are reasons to deviate from neutrality by favouring some activities over others. Deviations can be justified when the benefits of a simplification (such as removing many firms from VAT) outweigh the costs of added distortions or as a mechanism to meet redistributive goals. Some distortions arise from a desire to be internationally competitive and may be a price worth paying to ensure that the UK is an attractive location for multinational investments. But, too often, the ways in which our tax system influences industrial structure arise from an ill-thought-out desire to boost activities that sound like the kinds of things we must want more of.
A more structured way to design tax policies that will influence the industrial structure – and to evaluate the long list of tax breaks and special allowances – is to ask the following three questions.
Markets aren’t perfect or completely free. Imperfect competition is the norm and government intervention already shapes how markets operate. But while the many ways in which markets fail to produce the best possible outcomes lead to multiple arguments for deviating from a neutral tax system, doing so in practice is fraught with the dangers of complexity, special pleading and unintended consequences.
The first step when contemplating using policy to manipulate an outcome is to specify precisely what it is about current market outcomes that’s undesirable. What do we want more or less of, and why? Wanting more robots or small businesses or UK-made films because they’re ‘good’ in some way isn’t a sound enough rationale. If they’re good, why won’t the current market create the number of robots or films that balance the overall costs and benefits to society? If people aren’t willing to start a business without a subsidy, what makes us think that such a subsidiary should exist?
One specific reason that markets may fail to produce the socially optimal outcome is ‘spillovers’ (also called externalities). When individuals or companies make decisions, they don’t take account of the costs or benefits that their actions have on others; for example, they don’t value the lessons that others learn when a new technology is researched or a new business concept tried. Spillovers can mean that the market will produce too much of some things (like pollution) and not enough of others (like R&D) and this can be the basis for possible government intervention.
Knowing that there is an outcome we’d like to change is only the first step. There are rationales for the government to try, for example, to reduce pollution, problem drinking and childhood obesity while boosting R&D and entrepreneurship. Tax may not be the best tool, however. The government can also ban or regulate certain activities (such as advertising junk food to children), adjudicate competition (for example, between established firms and start-ups) and make direct investments (such as in science facilities or lending to new businesses).
Broadly, taxes are a good tool when changing prices (such as making the creation of pollution more expensive or R&D investments cheaper) or rates of return (such as making the post-tax rate of return higher if income is taken in capital gains than in earned income), and lead to more (or less) of the activity that the market is currently under (or over) producing. For example, R&D tax credits reduce the price of investing in research and thereby give companies an incentive to do more. If we want very specific outcomes (such as investment in malaria drugs), more direct intervention (for example, funding research or providing prizes) can be better targeted. And since we usually operate tax and direct measures simultaneously (for example, the UK spends around £3bn on R&D tax credits and about the same on direct support for business R&D), we should consider how they interact and what purpose different policy levers serve.
There may be a clear rationale for a policy and a reason why tax is, in principle, a good tool to use. But we still need to go to the effort of designing a policy that works and that doesn’t cause more problems than it solves. Consider entrepreneurship. One can justify considering policies that reduce the cost of trying a new business venture; however, that absolutely does not mean that a set of large, across the board tax breaks to all small business owners is a well targeted policy. It certainly might help to boost some of the activities that have externalities, but can also create a whole bunch of other unintended side effects (such as tax motivated incorporation) along the way.
In policy discussion, there’s a disappointing tendency to overestimate the ability to fine tune outcomes with tax and dramatically underestimate the set of unintended ways that people respond to tax incentives. Also, having identified that the market may underproduce entrepreneurs, it doesn’t mean that more is always good – we can have too many people starting new ventures. Finally, if the actual concern about start-ups relates to access to credit, then we should address that problem directly – offering juicy tax breaks for activities that generate capital gains doesn’t help a business that can’t get a loan today.
There are serious costs of sloppy thinking at the design stage. Poorly designed policies lead to inefficiencies (which ultimately make us poorer), unfairness (for example, when two very similar people get very different tax rates) and complexities. What’s more, since poorly designed policies often don’t work, we end up compounding distortions by layering new policies on top without considering how various policies interact.
One way to guard against a proliferation of poorly targeted policies is to evaluate what’s working. R&D tax credits offer a glimmer of hope here – there’s been a fair amount of research on their effects and there’s evidence that they lead to more R&D. This doesn’t mean the policy is perfect – plenty of the cost goes to investments that would have happened anyway and in an international context we should consider competition with other countries. But at least we have an evidence base. Most policies have never been evaluated. In many cases, there is no satisfactory way to do an evaluation because no data was collected and policies were introduced in ways that don’t allow their direct effects to be teased out.
While building industrial strategy into tax design might sound quite exciting, we shouldn’t get carried away. The bar for deviations from a neutral tax system (i.e. one that doesn’t affect our industrial structure) should be high. This is not because a neutral tax system is perfect but because attempts to use tax to help some groups too often create more problems than they are worth.
When we find problems or outcomes we’d like to change, we should first look at current policy. Removing current distortions is a seriously underrated policy option but is often much more sensible than layering another distortion on top. For example, there has been much concern about the effect of business rates on the high street, with proposed policy solutions including reliefs for pubs, retailers and small business and new taxes on online sellers. Instead, we could just fix business rates by moving the tax away from business buildings to land values, thereby removing all distortions to competition. This approach would not only make the system simpler rather than more complicated but it would actually work – whereas the quick fixes may offer some short-term relief but leave the structural issues in place.
If we were more willing to remove policies that weren’t working, we’d not only get a fairer, simpler and more efficient tax system but we could be more confident in experimenting with new policy options. If every new tax break or relief is destined to be a permanent fixture, we need to be much more mindful of how it will help to shape the long-term structure of the economy.
