- Take the poorest workers out of tax altogether by pegging the personal allowance and National Insurance threshold to the National Minimum Wage rate.
- Merge income tax and both employee and employer National Insurance contributions into a single system rate to boost transparency.
- Abolish all taxes on capital and transactions, such as corporation tax, capital gains tax and stamp duty.
- Broaden the VAT base and consider replacing income tax with a progressive consumption tax, so savings are not taxed.
- Reform business rates and council tax into a pure Land Value Tax on unimproved land values so that capital investment is not taxed.
Outside of impossible or improbable lump-sum taxation systems like taxes based on height, innate ability, or simply existence, and a small range of Pigovian taxes on externalities, there are no taxes which do not cause deadweight losses. That is, there are no real world taxes which simply move money around, without reducing economic activity at the same time. Once we’ve collected the revenue of well-designed congestion charges, fuel taxes, alcohol and smoking levies, and carbon taxes, we are always going to have to reduce economic welfare when we tax (even if we increase welfare overall by spending that money well). But that doesn’t mean that every way we raise our £700bn is equivalent.
Some taxes are costlier than others. Consider a consumption tax of some percentage on every good and service as a baseline. This reduces everyone’s buying power, and it distorts activity, since it makes every good more expensive with the exception of leisure, which is now cheaper (since every £1 of pay, and hence every hour of work, can buy less in goods). But other than that, it is neutral between different goods: everyone still spends their budget in the same proportion as they would do in the no-tax world. And everyone saves the same fraction of their budget as they would in the no-tax world.
Now consider a world with an income tax. In that world you pay tax not only on the fraction of your income you consume, but also on that fraction you save. Since this both reduces the principal upon which you earn interest, and the interest itself, directly, this makes future consumption more expensive, and saving (which provides the funds for investment) less attractive. In this world, not only do people do fewer hours and produce less output because consumption is more expensive, work is less well paid, and leisure is cheaper, but people also invest less, reducing future living standards. According to estimates from Nobelist economist Robert Lucas, the gains of scrapping all taxes on capital could be greater than those of eliminating the business cycle.
Generally, economists have concluded that tax systems are best when they rely more heavily on consumption taxes, and rely less on taxes that fall on capital or transactions. Great efficiencies could be obtained within the UK system without reducing revenues, including:
- Broadening the base of VAT to eliminate all exemptions and rate reductions—it would bring in £30bn, which could be used to eliminate less efficient levies, and compensate those who’d lose out (VAT is mildly regressive when considered within one year, and mildly progressive over the lifecycle)
- Scrapping stamp duty both on housing and on shares—the former reduces the efficiency of the housing market hugely, and makes housing shortages much costlier, since people are much more reluctant and less likely to move (as it costs so much more); the latter would raise the efficiency and information content of financial market prices, and make raising capital much cheaper (the tax ripples through the system, even though the rate is small in an absolute way)
- Merging employer NICs, employee NICs, and income tax, raise the thresholds to the income tax level—all of these, whoever hands them over, fall on the worker, and raising the threshold will deliver incentive-boosting income gains to those who need it most; clarity and simplicity are both desirable in a tax system as well
- Merge council tax and business rates, charge them at the same level, and charge them on unimproved land values—with modern econometric techniques, this is a technically trivial question; economically, there is no reason at all to subsidise one kind of land use, especially since the subsidy seems, in empirical work, to go to the landowner not the tenant (business or household).