Dr. Luke Redmond:

I will begin with the conversation on the gross value add, GVA. The GVA discussion focused on the figure used to underpin the productivity and output benefits associated with white-collar workers living in the intervention area but working outside it. In the updated model, it was originally proposed to use a 2016 GVA figure taken directly from the CSO but following conversations with colleagues at the Department of Public Expenditure and Reform, and in view of widespread conversations on the impact of multinationals in the economy, an issue was raised on whether the 2016 GVA figure might be slightly high and not totally reflective of the productivity levels in the domestic economy. In our discussions with the Department of Public Expenditure and Reform we agreed a different methodological approach. We stated we would go back to a 2011 GVA figure, which would then be scaled up by the rate of growth in modified gross national income, GNI*, over the 2011 to 2016 period.

The reason we went for this approach, and the Department of Public Expenditure and Reform highlighted this, was that in view of ongoing changes in national accounting practices modified GNI* was produced to give a more accurate view of the level of productivity in the economy. While multinationals are a very important contributor to the economy, the level of activity they produce can skew our true understanding of the domestic landscape. When we look at the calculation of GNI we have a GDP component and net factor income.

The reason I speak about this is because net factor income takes account of the income earned by Irish citizens and companies abroad from economic activity and property they hold. It also factors out income earned by foreign-owned companies for economic activity in an Irish context. GNI is a good figure but modified GNI was introduced by the CSO to take account of the activity of multinationals and specifically it takes account of the retained earnings of redomiciled foreign-owned companies in Ireland and the depreciation of foreign owned capital on national balance sheets. This meant that when we scaled the GVA figure from 2011 to 2016, we saw the growth rate of GNI* was 39% whereas across the same period the growth rate in GDP was 61%. We agreed with colleagues and used GNI* to give a more accurate figure than the productivity figure we were using across this period. It is interesting, and it speaks to my colleagues' points on the conservative approach we were taking.

When we talk about the productivity benefit that arrives to white-collar workers living in the intervention area, PwC considered it might be appropriate to use a GVA figure associated with white-collar workers, the assumption being these white-collar workers while living in the intervention area actually travel outside the intervention area to work in large urban areas. They undertake work in sectors that we would imagine have a higher value add relative to the national average. They work in professional services, IT and financial services. We felt this average value for white-collar workers would be a good GVA and could potentially be an option. However, on foot of ongoing debate with our colleagues in the Department of Public Expenditure and Reform as a result of the iterative process, it was concluded the GVA figure to be used was an average national GVA figure.

It included both white and blue-collar workers. The use of a GVA for white-collar workers would have led to a higher productivity uplift for the white-collar workers in the intervention area but who were not working there. The use of the national figure resulted in a reduction in the GVA. That was part of the robust approach we used.

The second area on which there was discussion was on the concept of displacement. When a policy intervention is introduced, such as the national broadband plan, and it leads to a positive uplift in economic activity or output in one sector of the economy, there is a risk that this might lead to a reduction in output in other sectors of the economy. The public spending code indicates that a cost-benefit analysis needs to provide some sort of estimate or indication of how this displacement effect has been taken into account. Officials in the Department of Public Expenditure and Reform have pointed out, and we agree, that there is a limit to domestic demand for output. There is an argument that any increased output in the economy from productivity improvements that the national broadband plan could be expected to deliver might lead to a fall-off in economic activity in other sectors of the economy. In the cost-benefit analysis we identify productivity and output improvements occurring at three levels, namely, for white-collar workers living in the intervention area but working outside it; for farm enterprises; and, for non-farm enterprises. The model took account of this through the application of a displacement factor onto the GVA or output uplift that the national broadband plan would deliver across these sectors.

The challenge here, and this is part of our robust methodology, is that there are no Ireland-specific estimates available for output displacement factors. To apply displacement factors to take account of displacement, we relied on a combination of international empirical research and an analysis of domestic data. The UK Government Department for Business, Innovation and Skills, now Business, Energy and Industrial Strategy, produced a research paper on research to assess the improvement of understanding additionality. It provides a range of displacement factors that can be applied to policy interventions across a range of themes and sub-themes and at regional and sub-regional levels. In the context of the farm and non-farm enterprise benefits, the cost-benefit analysis identifies these benefits emerging at a sub-regional level. We felt it was appropriate to utilise displacement factors from the UK Government paper at the sub-regional level. Specifically, the national broadband plan can be viewed as providing business development competitiveness support to the enterprises within these regions and the specific approach is almost an enterprise support. We applied an associated displacement factor of 16%. This was shared with our colleagues in the team in the Department of Public Expenditure and Reform.

Within the other principal sector, which is the big driver of productivity benefits relating to the productivity improvement for white-collar workers, the Department of Public Expenditure and Reform suggested we derive a displacement factor on the basis of looking at the volume of national output and identifying the split between the volume of that national output that goes to the domestic market versus the volume that goes to international destinations. An analysis of Central Statistics Office, CSO, data revealed it was difficult to show that split, so we looked at the total GVA produced in the economy and looked at the split between that produced by foreign-owned companies versus domestically owned, the assumption being that the volume of domestically produced GVA is for the domestic market and GVA produced by foreign companies would be for export, and that any change for domestic demand for output would not negatively impact that. That split showed that 60.3% of GVA is produced by foreign-owned companies, so we applied a displacement factor of 40% on the GVA uplift associated with white-collar workers. They were part of the iterative discussions with the team in Department of Public Expenditure and Reform. We had looked at using a displacement factor for white-collar workers of 20% based on an analysis of the UK Government research, but on discussion with Department of Public Expenditure and Reform, we felt the 40% figure was more appropriate. That was another contributor to the reduction in the benefit calculations.