The 2016 referendum held in the UK about the possibility to quit EU membership as well as a wave of populistic movements sweeping all over European Countries seem to suggest that less integration could be an outcome for the European Union. This paper has the aim to measure the cost of a missed integration, by highlighting what GDP growth would be in case of a missed integration. It does so by building a scenario of missed integration and compares it with a reference scenario. Scenarios are based on the Macroeconomics, Social, Sectoral, Territorial (MASST) model that has recently been updated to its fourth generation, whereby regional economic relations are tested econometrically. The estimated cause–effect chains are then the basis to build new scenarios simulated under complex sets of internally coherent assumptions in a simulation stage. The reference scenario presented is not a simple extrapolation of past trends; the post-crisis period registered structural changes to be taken into account for the future. In the integration scenario, we assume further integration within the EU to take place through the following changes: (1) higher trade flows among EU countries (“production integration effect”); (2) higher decrease in non-tariffs barriers (“proximity effect to larger markets”); (3) higher trust within and among countries (“social effect”); (4) higher quality of government (“institutional effect”); (5) stronger cooperation networks among cities (“cooperation effect”); and (6) higher exports (“market integration effect”). Results show that a more integrated scenario leads to faster economic growth across all EU countries. Territorial disparities are also initially lower in the case of more integration, although this difference abates over time. Lastly, the gains from integration are not spatially even and some regions gain more than others.

1. 1.

   On forecasting methodologies see, among others, Armstrong (1985), Hawkins (2001), Hendry and Clements (2001), and Loomis and Cox (2000).

2. 2.

   On foresight methodologies see, among others, CEC–European Commission (2004), Miles and Keenan (2000), and United Nations Industrial Development Organisation (2004).

3. 3.

   This work has been carried out before the 2020 COVID-19 pandemic, and therefore does not take its economic consequences into account. However, the aim of the paper remains valid. In fact, under the realistic assumption that the new crisis has to be taken into consideration in both the reference and the integration scenarios, its existence does not affect the relative results.

4. 4.

   Once again, while precise forecasts are not yet available at the time this is being written, the medical emergency due to the pandemic diffusion of the COVID-19 virus in the first half of 2020, and the ensuing lockdown measures taken in many Countries is likely to cause an even worse contraction of GDP in many EU countries than the 2007/2008 crisis. In fact, the IMF presently foresees a likely contraction of world GDP in 2021 for the first time in decades (World Economic Forum2020).

5. 5.

   The cluster analysis has been performed on the basis of the k-means method setting the target to obtaining three groups of Countries. Dissimilarity across groups has been defined in terms of Euclidean distance (Minkowski with argument 2). Lastly, centers of cluster have been identified with the first k observations from those to be clustered.

6. 6.

   Our working definition of the end of the crisis sets it to 2012, the first year in which EU GDP resumed pre-2008 levels. This does not imply that in 2012 all EU Countries achieved this goal; in fact, in purchasing power standard per capita terms, in 2012 Greece, Spain, Croatia, Italy, Cyprus, the Netherlands, Portugal, Slovenia, Finland, and the UK were still off target.

7. 7.

   It here suffices to mention that inside the Old 15 group, the best performance in terms of GDP growth is found for Luxembourg, Belgium, Denmark, The Netherlands and Austria, while among CEECs top performers include Estonia, Slovakia, Bulgaria, Hungary, and Lithuania.

8. 8.

   See Rodrik (2018) for a thorough review of the literature on trade openness.

9. 9.

   UK’s percentage gain w.r.t. the EU is obtained as the ratio + 11%/+25%.

The scenario exercise was in a first version being developed within the ESPON ETRF project. The authors would like to thank Prof. Roberto Camagni, Politecnico di Milano, for his advice during the development of the project. The Authors would also like to thank Prof. Barbara Chizzolini, Bocconi University, for helping with the coding of the new version of the MASST model, and Elisa Panzera for skillful research assistance with the analysis of data described in Sect.1.3. All remaining errors are our own.

Authors and Affiliations

1. ABC Department, Politecnico di Milano, Milan, Italy

   Roberta Capello & Andrea Caragliu

Corresponding author

Correspondence toAndrea Caragliu.

Editors and Affiliations

1. Hokkai-Gakuen University, Sapporo, Japan

   Soushi Suzuki

2. University of Bologna, Bologna, Italy

   Roberto Patuelli

Annex 1

Reprints and permissions

© 2021 Springer Nature Singapore Pte Ltd.

Policies and ethics