New ESRI research compares productivity levels in Ireland and Northern Ireland

The Economic and Social Research Institute in partnership with the Shared Island unit in the Department of the Taoiseach have produced and published the first comprehensive study comparing productivity in Ireland and Northern Ireland.



The research examines trends in per worker productivity across both regions, as well as assessing and modelling the drivers of productivity in both jurisdictions. The main findings from the research are:

  • Examining trends at the level of the economy, productivity levels in the two regions were broadly equivalent in 2000. Over the period 2001 to 2020, productivity levels in Ireland have trended slightly upwards, while in Northern Ireland productivity levels have been trending downwards. By 2020, productivity per worker was approximately 40 per cent higher in Ireland compared to Northern Ireland.
  • The research exploits annual sectoral level data on labour productivity, measured by gross value added per worker, and finds that of the 17 sectors for which comparable data is available, productivity levels in Ireland noticeably exceed those of Northern Ireland in 14 sectors, with particularly large gaps in Administrative and support services activities; Financial and insurance activities; Legal and accounting activities; and Scientific research and development. Northern Ireland has an advantage in the two sectors of Electricity and gas supply and Construction.
  • The model for Irish productivity shows that sectoral productivity increases with the employment share of educated workers and levels of investment. For example, with respect to education, the model suggests that a 1 per cent increase in the share of graduates employed generates a 1 per cent increase in sectoral productivity. The report also finds that export intensity is an important factor in driving Irish productivity.
  • Despite using comparable data sources and the same estimation method, we find no evidence of a relationship between the range of factors captured in the model (education, investment, exports etc) and Northern Ireland productivity. The models for Northern Ireland do not show significant results for the usual drivers of productivity.
  • A separate modelling approach (a decomposition) is used to measure the extent to which differences in productivity levels in the two jurisdictions are explained by differences in their amount of factors, such as educated workers or investment, that should ultimately determine productivity and other factors. If the Irish productivity model is re-estimated using Northern Ireland endowment levels, the analysis shows that all of the observed productivity gap can be explained by differences in investment and labour market related endowments (in particular lower proportions of workers educated to post-secondary level in Northern Ireland) between the two regions.
  • The analysis points to the need to rapidly expand investment and improve skills in Northern Ireland, particularly at the post-secondary level. However, our models also suggest that without a comprehensive strategy aimed at improving competitiveness among Northern Ireland firms, the reform of education and skills provision and increasing investment in isolation are not guaranteed to enhance Northern Ireland's productivity. 

Adele Bergin, one of the report authors said, “While some existing studies have compared Northern Ireland’s productivity to that of regions in Great Britain, we believe that this is the first comprehensive study to compare productivity in Ireland and Northern Ireland. The research shows a widening productivity gap between Ireland and Northern Ireland; with productivity per worker being approximately 40 per cent higher in Ireland compared to Northern Ireland in 2020. The usual factors (education, investment, exports) were found to be important determinants of Irish productivity. However, no significant relationship is found between these factors and Northern Irish productivity. It may be the case that there are other factors that need to be considered in devising a policy response.”