Economics: The recent release of initial results from the 2006 Census showed that around 400,000 foreign nationals now live in Ireland, an amount equal to 10 per cent of the population.

While this percentage is not exceptional when judged relative to other EU countries, the speed with which the proportion has grown in Ireland is noteworthy.

It has also meant that research on the impacts of immigration is lagging behind the phenomenon itself.

Even so, we do have some results from economic research which inform our understanding. What does this research tell us about the impact of immigration and what does it miss?

A simple example illustrates some of the problems involved in measuring the impacts of immigration. Suppose a country is growing strongly, with unemployment falling and wages rising. Such a country is likely to attract immigrants. Indeed, as growth accelerates in the country, larger numbers of immigrants are likely to arrive.

If a researcher were to take the data for this country, they would find a positive correlation between economic growth and immigration. However, they would clearly be wrong to conclude that the immigration was causing the growth.

What is more, if the researcher observed that wages continued to rise as immigrants arrived, they would be wrong to conclude that immigrants were not depressing wages. Wages could have risen faster if the immigrants had not arrived.

The crucial point that this example illustrates is that you cannot measure the impact of immigration just by looking at a country before and after the immigration inflow. You need to compare the country after the arrival of the immigrants with what the country would have looked like had the immigrants not arrived.

In order to do this, you need to use a statistically constructed model of the economy that allows you to simulate the "without immigrants" scenario.

In a paper published earlier this year, some colleagues and I explored the issue of immigrant impacts using the simulation approach. We found that an inflow of 70,000 immigrants (the amount who arrived in Ireland in 2005) would, on average, lower wages by about 4.5 per cent and increase GNP by about 3 per cent.

It is important to stress that we were not arguing that wages actually fell by this amount. We were saying that wages were lower than they otherwise would have been. It is also important to note that, while GNP would have increased by 3 per cent, the increase in GNP per head would be lower at around 1 per cent.

The reason for the difference is that some of the increase in national output accrues to the immigrants themselves.

This is an important distinction because many would argue that economic policy should aim to increase GNP per head (and not GNP per se) because this is a better measure of living standards.

One of the crucial features of immigration is that inflows leave some people better off and some people worse off. I have already noted that wages were found to be 4.5 per cent lower on average than they would otherwise have been as a result of immigration but this figure hides some important differences across groups.

If Ireland's well-educated immigrants were employed in jobs that matched their high levels of education, it was the wages of the high-skilled residents that were reduced.

The wages of low-skilled existing residents would actually increase in this scenario because spending by the larger high-skilled workforce would increase the demand for low-skilled labour.

The overall result from this analysis was that immigration was playing a role in increasing GNP. It was achieving this by ensuring that the increased demand for labour was being met by an increased supply. Had the supply not been forthcoming, wages would have risen more quickly thereby choking off the increased demand. In this way, although the impact of immigration on GNP was positive, employees lost out through wages being lower than they otherwise would have been.

While this research produced useful information on the short-run impact of immigration, it leaves many questions unanswered on the longer-run impact.

Some economists argue that, although the short-run impact is for immigrants to depress wages and/or to increase unemployment, the longer-run effect is positive for both GNP and wages.

Examples of these longer-run impacts include: increased dynamism in the labour market as a result of the positive characteristics of immigrants (such as entrepreneurship and ambition); diversity in the workforce leads to greater innovation; greater competition in the labour market leads to increased effort.

While all of these factors have been well integrated into economic theories of immigration, their nature makes it difficult to produce empirical evidence in support of the claims. For example, it is difficult to measure characteristics such as ambition or how one person's ambitious nature can have a positive impact on others.

Some argue that the buoyant economic performance of the US economy over the last 10 years relative to the EU is evidence of the long-run positive impact of immigrants. While this might be true, it has not been shown empirically.

Although the analysis we published earlier this year was focused on the short-run, its findings are important.

By identifying and quantifying the initial impacts of immigration, we showed how an overly liberal policy on immigration could lead to downward pressure on wages at a rate that would have been viewed as disruptive, even in the context of positive GNP impacts.

However, the potential longer-run impacts of immigration point to the potential value of a continuing, moderate rate of inflow.

• Alan Barrett is senior research officer at the ESRI