Marije Kooistra (p), Richard Rijnks, Aleid Brouwer

Susana Franco

Beyond-GDP frameworks are gaining traction as policymakers increasingly prioritize well-being and quality of life over material wealth. While research has explored the relationship between well-being domains and life satisfaction, their relative importance is often unknown (Clark, 2018; Samavati & Veenhoven, 2024). This study addresses this gap by assessing how different well-being domains contribute to life satisfaction across the distribution. Using the broad prosperity framework commonly used in Dutch policymaking, we analyze the relative importance of seven domains (CBS, 2023). To account for variation across the distribution, we employ quantile regression and stratify by age groups. Additionally, we make a methodological contribution by applying Gradient Boosting, a machine learning approach which handles non-linearity and multicollinearity well and was shown to be promising for wellbeing research (Oparina et al., 2025).

Using 2024 survey data (n=5409) from two Dutch provinces, we found that some domains are indeed more important than others. There is fairly limited variation in the order of domain importance across age groups and across the distribution of life satisfaction and that GBQR (Gradient Boosting Quantile Regression) has both advantages and disadvantages compared to linear regression in explaining life satisfaction.

Our results indicate that some domains are consistently more important than others, with health—particularly mental health—emerging as the most significant factor. This highlights the necessity of accessible mental health services. Material wealth ranks fifth in importance, following labor and leisure (which include satisfaction with work, daily activities and free time), social cohesion, and housing. The remaining domains each explain less than 5% of the variance in life satisfaction. Domain importance is consistent across quantiles, meaning that in general what is important for those at the lower end of the distribution is no difference in importance from those at the top end. Overall, the GBQR model preforms well with the pseudo R-squared exceeding 0.5 at some quantiles. At the extreme quantiles it is, however, outperformed by linear regression. The models for individuals under 35 and those aged 35–55 exhibit the highest explained variance.

Our main contribution is the improved understanding of the dynamics of wellbeing frameworks. While it provides empirical backing for domain weighting in wellbeing monitoring frameworks, it also provides arguments against this practice.  Variations in relative importance suggest that weighting domains based on mean importance may misrepresent certain groups, particularly younger individuals.

Susana Franco (p), James R. Wilson

Eveline van Leeuwen

This paper presents the process of constructing a framework for territorial competitiveness for wellbeing, as a collective endeavour of a research institute that has the mission of fostering the competitiveness of the region where is based through action-oriented research to improve the well-being of their citizens. The conceptualisation phase of the process has involved an iterative process to visually reflect the concept and to reduce its components. The resulting framework contains three distinct elements: (i) dimensions of territorial performance, disaggregated in economic/business performance, which typically forms part of the analysis of territorial competitiveness, and wellbeing performance, which should constitute the ultimate goal of territorial competitiveness; (ii) the dynamic levers of competitiveness for the territory, which include all the factors that have some impact on the territory’s competitiveness and wellbeing, and (iii) the structural context of the territory, which includes characteristics that are largely given, change slowly over time, and are neither good nor bad per se. Completing the framework are two cross-cutting dimensions: inclusiveness and international connection. The implementation phase is ongoing, starting from a selection of indicators that were initially analysed in 2021 and incorporating different indicators in subsequent years that allow a deeper understanding of various parts of the framework through different lenses.

Eveline van Leeuwen (p), Aleid Brouwer

Marije Kooistra

Economic growth has been a dominant goal and narrative, basically since the development of GDP after the Second World War. The modern concept of Gross Domestic Product (GDP) was first developed by Simon Kuznets for a 1934 U.S. Congress report, where he warned against its use as a measure of welfare. After the Bretton Woods Conference in 1944, GDP became the main tool for measuring a country's economy. And it not only became the main measuring tool, economic growth (as measured by GDP) also became the ‘main solution for everything’: from poverty, unemployment, environmental destruction, climate change, financial instability, etc. (Daly, 2019). Many scientists, including Noble prize winner Joseph Stiglitz, have shown that society in general and the economics discipline more specifically needs a different compass than GDP (Hoekstra, 2019). Alternative economic models have been proposed, such as Doughnut Economics, Sustainable Development Goals, Steady State Economics, De-growth and Green Growth (D’Alessandro et al., 2020). The fundamental value which these models have in common is that they make a distinction between present wellbeing, future wellbeing (sustainability) and also the distribution of wellbeing within and between countries (inclusion). But how to conceptualize and measure this in a way that is meaningful (i.e. measuring what matters) and clear (through a moderate number of indicators), as well as appropriate for the right spatial scale?

In this paper, we analyze how economic growth is used within the regional science literature: to which topics is it linked, what conclusions are drawn and how is this translated into policy recommendations? In a next step, we will analyze to what extent new concepts as broad prosperity and postgrowth are used and implemented.
To answer these questions, a structured literature review within the ERSA/RSAI journals will be performed.