Michael Keen, Deputy Director of the IMF's Fiscal Affairs Department, will be the keynote speaker at this year's WIDER Development Conference on Public Economics for Development (/node/62695). Back in 2014 we interviewed him at our offices in Helsinki, and asked him about his work on poverty, tax, and spending.
Why general subsidies are not a good way to help the poor
In the interview, Keen acknowledges that subsidies on things such as food and petroleum might seem like a tempting way to reach the poor. After all, the poor spend a larger proportion of their disposable income on these things than the rich, and it seems sensible to see these policies as pro-poor in nature.
However, Keen points out, the rich spend a larger absolute amount on food and petroleum, and most other products that might be subject to subsidies. As a result, most of the benefit in absolute terms goes to the rich, not the poor. Keen's research shows that for every 100 dollars a government forgoes in revenue for a food subsidy, the poorest get 4 dollars of the benefit. The question for Keen then is, are there better ways to use that 100 dollars to help the poor?
Alternatives to subsidies
In advanced economies with established structures of social support the answer can be relatively straightforward. Governments can simply choose not to forgo the 100 dollars, and use it to bolster social transfers to the poorest in society.
Keen accepts that this may be more difficult to do in developing and emerging economies where social support systems may not have the capacity to reach the poor effectively. However he argues that even the very blunt tools of redistribution available in these countries are often better for reaching the poor than reduced taxes and subsidies.
Keen also suggests rethinking the common view that VAT is a regressive tax. While it may be true that VAT in itself is regressive, if the revenue raised is spent on pro-poor measures it is not necessarily the case.
Keen also highlights property tax as a good example of a tax where there are potential gains in both fairness and efficiency terms. Even if we leave aside the potential redistribution effect of a property tax it can be quite an efficient tax base, especially in developing countries where tax collection capacity is weak, because property is more difficult to hide than income.
Practical matters often act as barriers to research
Keen argues that to a large extent practical issues are the barrier preventing...