Today's "Dinheiro Vivo" supplemento to Jornal de Noticias e Diario de Noticias had an essay by me on fiscal devaluations. (It is here, but a better version is here.)
There are some pros and cons to this measure, but many of the arguments that have dominated the public debate in the last month are, in my view, not valid. In particular:
1) Fiscal devaluations have nothing to add to our deficit target, positive or negative. They are budget neutral. They are a growth measure.
2) Devaluations are not some funky idea by esoteric or disreputable economists or crazy wonks. They are part of almost every IMF plan, including those in Portugal in the early 80s.
3) The price of Portuguese non-tradables does not change with the measure. The price of imports rises, the price of exports falls, and that is it.
4) Profits are likewise unchanged.
5) Fiscal devaluation is not the same as lowering payroll taxes. Raising VAT is as important as lowering the payroll tax in the measure
6) One feature that hasn't been given its due attention is that you have to raise the consumption tax on sectors exempt from VAT, especially real estate.
7) According to the study put out by the Bank of Portugal and several Ministries, lowering payroll taxes by 3.7% as part of a fiscal devaluation leads to a fall in the real exchange rate of about 2.2%. That is a lot. The measure has a lot of bite on competitiveness, in spite of what a naive look at the labor content of exports may show.
8) Contrary to what the press reported, that official study, in my view, supports the use of a fiscal devaluation. The study spent (too) much of its time talking about lowering payroll taxes, which is not the same as a fiscal devaluation (point 5 above). Also, its more reliable estimates show a positive impact of 0.5% of GDP to a 3.7% lower TSU plus 2.2% higher VAT combination. That's a big impact. Extrapolating, it says that if you doubled or quadrupled the measure (like I have defended, and the IMF has supported) you could get a huge impact, likely well above 1% of GDP. Show me any other measure that has that short-run impact on output.
