In my previous post, I suggest one possible rationale for levying a VAT: if you think that the US needs to raise more tax revenue, and if you think that political opposition to significant increases in income taxes is strong, then adding a new VAT on top of existing income taxes is one important option to consider. Let me take this chance to go into the pros and cons of a VAT in more detail. (For a brief primer on what a VAT is, you might want to see Wikipedia, or Bruce Bartlett’s latest post.)
Advantages of a VAT:
- It can raise lots of new tax revenue with relatively low tax rates. A reasonably broad-based 10% VAT in the US would raise several hundred billion dollars of tax revenue. (Obviously, this is only an advantage if you think that we need to raise more tax revenue in the first place.) Raising the equivalent revenue through income taxes would require far higher marginal tax rates.
- Tax evasion is a relatively small problem. Because a VAT requires each step in the production process to report what the previous step in the production process sold to them, there’s a degree of automatic enforcement in VAT administration, unlike with sales or income taxes. Furthermore, since a VAT is so broad-based, the rate can be set very low, which means that each individual business has relatively little incentive to try to cheat. The experience of countries with VAT is that compliance is higher than for most other types of taxes (much higher than for sales taxes, for example), and the costs of enforcement are lower.
Disadvantages of a VAT:
- As is the case of all consumption taxes, a VAT is regressive; lower-income people pay a larger fraction of their income in taxes than higher-income people do.
- Since a VAT taxes consumption rather than income, it hits those who consume but don’t earn income harder, e.g. retirees.
- A VAT may alter international competitiveness.
The advantages of imposing a VAT are fairly straightforward. I tend to think that the US government will indeed need to raise more revenues as a fraction of GDP in future decades, and that a VAT would be a less painful way of doing so than most alternatives that I can think of.
Sure, there are other ways to raise revenue; so the test is whether the VAT does it more efficiently (i.e. imposing fewer economic distortions and creating fewer incentives to cheat) and more equitably than the possible alternatives. Regarding efficiency, the answer seems pretty clear. That’s why every other developed country around the world has concluded that the VAT is indeed a better way to raise the additional needed revenue. Equity, however, is another story…
So let’s get to the problems with a VAT. They are potentially substantial, and deserve careful scrutiny. When AB first started this blog, he devoted several posts to explaining the regressivity problems of consumption taxes. To get a good overview of those problems, take a look at his posts under the topic heading “Consumption Taxes” in the left-hand column of this blog.
Put simply, consumption taxes, including a VAT, are regressive. It’s hard to get around that fact. But there are a couple of things that you can do to mitigate these effects. For example, you can exempt certain items that poor people spend a larger fraction of their income on (food is a popular example). Even better, however, would be to couple the imposition of a VAT with changes in income taxes to make them more progressive.
This is a crucial point: a VAT by itself is surely regressive; but a VAT coupled with income taxes does not have to be. As an example, simply note that all European countries have a VAT. Yet all of them have a total net tax system that is far more progressive than the US’s tax system. I’ll have more to say about this in another post that I’m working on.
The potential problem of the VAT hitting retirees harder is probably less of a problem than one might think. First of all, since the “fiscal nightmare” is almost completely the result of health spending on retirees, if they ended up paying a higher share of the VAT than others that might only be fair. But secondly, the disproportionate impact on retirees is likely to be substantially mitigated in the first place. A VAT has the effect of raising overall price levels in the economy, as businesses incorporate the VAT in their sales price. But that means that the cost of living adjustment that retirees receive on their Social Security benefits would also go up by the same amount. The net effect is a wash for those retirees who depend primarily on their Social Security benefits as income.
The last potential drawback of a VAT also turns out not to be a problem. Economic theory suggests that a “destination VAT” – that is, a VAT imposed on all goods, but only those goods, consumed in the country – should not affect international competitiveness one way or another, in and of itself. (Note that a destination VAT requires that all imports are taxed at the regular VAT rate, and that exporters receive a rebate on all VAT taxes that the exported item has incurred during production.) As a slight twist on this result, note that if a VAT falls more heavily on traded goods than non-traded goods due to some items being exempted, then a VAT would be expected to cause a fall in both imports and exports. For a paper explaining the logic, see the classic Krugman and Feldstein paper “International Trade Effects of Value Added Taxation”.
The empirical evidence seems to confirm the economic theory. A paper by Mihir Desai and James Hines suggests that the VAT does have a small effect on trade, but that it simultaneously reduces both imports and exports with no substantial net effects. On the other hand, an even more recent paper by the IMF, “Domestic Taxes and International Trade,” was unable to find any effect of the VAT on trade. Either way, the effects of a VAT on trade seem to be small and neutral, in the sense of not really changing the balance of trade or a country’s competitveness.
To sum up, I would therefore offer this assessment of a VAT: it has some distinct advantages, and it has a huge potential disadvantage (regressivity) that can be overcome with the right other policy changes. So personally, I could possibly get behind the idea of a VAT to raise more taxes… but only if combined with changes to the income tax code to make them significantly more progressive.
But before we do that, my preferred solution to the fiscal nightmare would be this: enact a sweeping change in how health care is paid for, and adopt single-payer health insurance. Give it 10 years, and see how health care spending and the growth in spending adjust. I suspect that such a change by itself would solve much or all of the fiscal problem. If there’s still a fiscal problem after that (and there might be, I realize), then consider a VAT – but only in tandem with changes in income taxes to make them more progressive. The fact that every developed country in the world has a more rational health care system and a more progressive tax code, and also has a VAT, may be a sign in the VAT’s favor.