Wednesday 9 November 2011

Fiscal devaluation, a PR failure?

The memorandum of understanding signed last May by the main Portuguese political parties contains a measure designed to improve competitiveness and accelerate the necessary improvement in the trade balance. The measure consists in a budget neutral tax swap: an increase in the effective VAT rate (by increasing the low and intermediate rates) and a reduction of the social security tax paid by employers (TSU). The measure has been compared by scholars, journalists, experts and others (myself) to a nominal devaluation. The idea was that to show that even if the euro has eliminated the exchange rate, euro members could implement policies with a similar outcome to realignments of the exchange rate.

This comparison might have been unproductive for two reasons.

The first is a public relations failure. The name of the suggested policy contains the word devaluation. I remember an influential journalist calling the fiscal devaluation a "new drug" to replace the old one (i.e. nominal devaluation).
While from a from a technical point of view this affirmation is incorrect, it reminds us that "devaluation" is considered a dirty word. In fact, southern euro countries have adopted the euro in part to tie their hands and avoid external adjustments through devaluations.

The second is that for practical (as opposed to academic) purposes we should have compared the fiscal devaluation with a competitive disinflation. The latter is the market based mechanism that takes place within a fixed exchange rate system to adjust to external imbalances (in the absence of fiscal transfers and with limited labor mobility). It can be described as follows: unemployment increases and pushes the growth rate of nominal wages down until the country’s competitiveness is restored. It is obviously a painful mechanism that relies on two key market transmission mechanisms:

1) the decrease in nominal wages in face of higher unemployment, and
2) the decrease in prices in face of a decrease in the wages.

Consider impediments to the first mechanism: when nominal wages are sticky and adjust slowly, the competitive disinflation requires a high unemployment rate. Are nominal wages (downward) flexible in Portugal? If I remember well it is very difficult (unconstitutional) to decrease private wages. The Portuguese government has recently cut the public sector wages. This should help alleviate the necessary increase in unemployment as it can signals to the private sector wage setters the necessity to accept a lower wage.

Now consider the alternative of a fiscal devaluation. The decrease in the social security contributions aims at reducing the cost of labor in substitution of a decrease in the nominal wages. It does not require unemployment and works when wages are sticky. Moreover it affects simultaneously both public and private sectors.

It appears that on the first key transmission mechanism, fiscal devaluation works better than competitive disinflation. For what regards the transmission from labor costs to prices, there are no differences between the fiscal devaluation and the competitive disinflation.

A critique to the decrease in the TSU is that it will lower revenues and jeopardize the financial stability of the social security system. The fiscal devaluation is designed to be budget neutral: the second element of the measure is an increase in consumption taxes that offsets the decrease in revenues. In reality, to increase consumption taxes in Portugal is objectively difficult as VAT has already been increased to diminish the budget deficit. Again the comparison with the competitive disinflation is useful. A decrease in nominal wages decreases social security contributions and requires a decrease in benefits to maintain financial sustainability. This suggests that the loss in revenues that follows the substantial decrease in the TSU can be financed through a combination of lower benefits and higher consumption taxes. To a first approximation if wages are expected to decrease by 25%, social security contributions will decrease by 25%. To maintain the solvency of the social security system benefits will have to decrease by 25%. The same reduction of benefits could be used to finance part of the decrease in TSU so that the increase in the consumption taxes can be lower.

Another advantage of the fiscal devaluation is that the real debt due by wage earners (debt deflated by wages) does not increase. This is not a secondary point in an environment of fragile banks balance sheets.

The fiscal devaluation and the competitive disinflation have also different distributional effects. The latter are important as they are likely to determine the political feasibility of the measure.

I am surely missing something, but as of today I see the fiscal devaluation as a superior policy than trying to manipulate the nominal wages in the private sector or to wait for unemployment to be sufficiently high to exert downward pressures on those same wages. So what explains the opposition to such a measure? Should we call the fiscal devaluation "fiscal disinflation"?


  1. I would answer your question with two ideas that you presented:
    1) no one is talking about the alternative way, the increase in unemployment and the decline in nominal wages
    2) the VAT has already been increased sharply, so a new and painful increase would be needed.

    Another idea, that you are missing, is the fear that in the non tradable sector, where competition is weak, a decrease in costs will only translate into higher profits.


    Why merely REDUCE payroll tax? Why not ABOLISH it altogether? If the revenue were replaced by VAT, domestic retail prices of domestic products need not rise; embedded payroll tax would simply be replaced by embedded VAT. Of course retail prices of IMPORTS would rise due to the higher VAT. But domestic consumers would be compensated by better employment opportunities due to abolition of payroll tax.

    And why stop at payroll tax? The cost of labour for employers can be reduced further -- without reducing employees' "take-home" pay or widening after-tax wage inequalities -- by replacing Pay-As-You-Go personal income tax IN THE HANDS OF EMPLOYERS. Employees would continue to receive credit for the PAYG tax withheld by their employers (calculated on their "grossed-up" wages). But businesses, instead of forwarding the withheld PAYG tax to the government, would pay a higher VAT. In the aggregate, the withheld PAYG tax would cover the additional VAT, with no need to raise domestic retail prices of domestic products.

    Three advantages would follow from this combination. First, payroll tax and PAYG personal income tax would be completely removed from the marginal cost of labour as seen by employers. That means more jobs, hence more domestic demand for domestic products. Second, the additional VAT, unlike the tax component of the cost of labour, would not feed into export prices. That means more foreign demand for domestic products. Third, because the extra jobs would reduce welfare spending, not all of the lost revenue from PAYG tax would need to be replaced. That would allow a slight FALL in prices of domestic products.

    I have suggested this replacement of payroll tax and PAYG tax as a lifeline for indebted nations on the edge of the eurozone -- in particular, Ireland: .

  3. P.S.: Perhaps it is conceptually simpler to let employers "offset" payroll tax and PAYG personal income tax against VAT; see "Fiscal devaluation on steroids".