Fiscal Policy and International Competitiveness

Two current policy problems for Ireland are to tackle the loss of external competitiveness and to determine the appropriate level and composition of government spending.  These issues are linked, since government spending affects the real exchange rate for Ireland, through its impact on the relative price of nontraded goods in terms of traded goods.

In a new paper  Fiscal Policy and International Competitiveness: Evidence from Ireland(joint with my TCD colleague Vahagn Galstyan), we show that the long-run behaviour of the real exchange rate and the relative price of nontradables is increasing in the long-run level of government consumption but decreasing in the long-run level of government investment.

The intuition is that government consumption tends to drive up economy-wide wages and nontraded prices (since the public sector competes for scarce labour and non-traded inputs), while government investment in the long run improves productivity (especially in the non-traded sector) which is associated with a reduction in the relative price level.

The appropriate levels of government consumption and government investment depend on a range of socio-political factors, but these results are worth noting in any debate about the connections between fiscal policy and external competitiveness.

6 replies on “Fiscal Policy and International Competitiveness”

It makes sense.

For your next trick, can you decompose this investment expenditure further, to tell us what sorts of investment is most useful? I guess investment boosting non-traded sector productivity, but what does that mean in practice? Is transport infrastructure an input to the non-traded or the traded sector? Broadband? etc?

I guess another big factor not modelled here, but which you implictly acknowledge in your comments above, is how labour markets respond to government expenditure (and, hence, taxation). The Celtic Tiger model is that higher taxes require higher wages and hence a higher relative price of non-tradables. On the other hand, the experience in other European economies in the 1950s and 1960s was that government safety nets of one sort or another compensated workers for lower wages. Complicated stuff.

Philip, did you try splitting government consumption into compensation and purchases of goods/services? Also, would it matter if the change in compensation was driven by wages per employee versus number of employees? I ask because it may be relevant for the current debate on public sector reform and because I used to work with someone who was convinced that cuts in wages per public sector worker delivered much more disinflation than cuts in worker numbers (with average wages constant).

Moving outside Philip’s model, I would say that holding the wage bill wL fixed, you would get a lower relative price of non-tradables with a lower w and higher L. I am thinking of what wages Irish workers will accept. First, keeping wL fixed, taxes are fixed, and so any compensation workers need on that score is fixed also. So we can forget abou that mechanism. Second, if the public and private labour markets are linked (OK, perhaps doubtful) then a lower public sector wage will feed into lower private sector wages, and hence lower non-traded goods prices. Third, a higher L will mean more services, and like Dani Rodrik I tend to believe that when times get uncertain and workers worry about risks, certain government services can help lower the risks workers face, which might make them more sanguine about lower wages.

(On the other hand, you could argue that a few high paid public workers will blow all their wages on imports, skiing holidays etc, while lots of lower paid workers would spend money on more domestic goods…Depends on income effects.)

And furthermore, holding wL fixed, if you lower L, you lose tax revenue and have to pay people on the dole, thus raising required tax rates. Which strengthens the case.

Indeed. Though I would have thought that with a progressive tax system, lower L with wL fixed would give higher tax revenues. Dole payments would rise, but higher unemployment should put downward pressure on private sector wages. So the wage differential gets even larger.

Comments are closed.