For a long time, debates about secession focused on separatist conflicts in developing countries (Collier and Hoeffler 2006, Morelli and Rohner 2015). Recently, however, secession is increasingly becoming an urgent issue in Western democracies. Following the unofficial 2017 independence referendum in Catalonia, tensions between the region and its national government dominate Spanish politics. After the (supposedly final) Brexit decision, Scottish National Party (SNP) leader Nicola Sturgeon did not hesitate to call for a second referendum on independence from the UK. How can economic research help us understand the current surge of secessionism?
Economic size-of-nations models (e.g. Alesina and Spolaore 1997) consider how regions weigh the advantages of being part of a greater state against the disadvantages. Economic benefits of larger states include improved trade opportunities and a more efficient provision of public goods. For instance, a joint national defence is much more affordable than individual regional armies are.
Political autonomy is commonly regarded as the main advantage of independence. This is more relevant the more the culture and preferences of a region differ from the rest of the country.
Existing models, however, miss one key to understanding the electoral success of secessionist parties: regional resources, the regional wealth and revenues that they create, and their redistribution through fiscal transfer systems. This redistribution clearly plays an important role in the rhetoric of successful secessionist parties. A good example is how the New Flemish Alliance in Belgium highlights that “wealthy Flanders should not be subsidising poorer Wallonia, whose regional government is alleged to be wasting money”.1
In our recent study (Gehring and Schneider 2020), we show that integrating regional resources systematically in a formal economic model yields some important insights. First, what matters for secession support is not regional resource wealth in absolute terms, but relative to other regions. Second, changes in regional resource wealth only play a role when there is baseline support for secession – for instance, due to cultural differences, historical repression (Dehdari and Gehring 2019), or other tensions with the central government. In general, a region that is itself at the centre of political power – like England in the UK – has little incentive to leave this constellation.
Resources, regional wealth, and electoral support for secession
The wealth of a region stems not only from natural resources; any resource that generates revenues and can be taxed by governments is a source of regional wealth. This also includes geographic characteristics, deep-rooted institutions, and region-specific human capital. For example, secessionist claims in Istria in Croatia are based on the revenues from thousands of tourists flooding the region’s pristine beaches every summer. Northern Italian separatism focused on wealthy areas with well-functioning institutions.
Changes in relative resource value often seem to foster support for secession. In Belgium, the advancing globalisation and growing importance of Antwerp as a trading hub benefited Dutch-speaking Flanders, while the economy in French-speaking Wallonia was in decline. Those changes in relative wealth overlap clearly with the increased electoral success of Flemish separatist parties. Or consider Greenland: currently associated with Denmark and depending on Danish support for large parts of its budget, some first resource discoveries – with the prospect of more following the melting of the ice caps – led to a landslide win for a 2008 referendum on more autonomy. Following a major decline in oil prices, however, plans for secession were ‘put on ice’.
Because of the diversity of resources across regions, there is little sense in directly estimating the average effect of resources. Instead, we begin our empirical analysis by examining changes in regions’ relative wealth over time, measured by GDP per capita. A 10% increase in relative regional wealth is associated with an increase of 3 percentage points in the vote share of secessionist parties. Hence, economic factors have a sizeable impact on support for separatist parties.
Quasi-experimental evidence from oil discoveries in the UK: Scotland versus Wales
Obviously, it takes more to make a compelling argument about the causal impact of regional resources. We exploit oil discoveries as a natural experiment to analyse secessionism in the UK. Our strategy is based on the existence of separatist parties in Wales and Scotland since the early 20th century: the Welsh Plaid Cymru and the Scottish SNP. The decisive difference that started to arise between the two regions is that almost all of the UK’s oil discoveries since the 1970s have been located within the maritime boundaries of a potentially independent Scotland (Figure 1).
Figure 1 Distribution of oil fields and Scottish maritime boundaries
Note: The map shows the hypothetical Scottish maritime boundaries as well as the locations of the oil fields discovered in the UK sector.
We use difference-in-differences to identify the causal effect of oil discoveries on the rise of the SNP. We assume that in the absence of the discoveries, vote shares of the two parties would have developed similarly. Figure 2 shows that, without clear economic arguments to base their campaigns on, the parties’ performances in UK general elections followed a nearly parallel trend, at very low levels of support. This changed dramatically when the first oil discoveries in the early 1970s caught parties and voters by surprise (MacKay and Mackay 1975).
Figure 2 Oil discoveries and electoral performances of the Scottish SNP and the Welsh Plaid Cymru
Notes: Panel A shows the temporal distribution of oil discoveries and giant oil discoveries (oil fields exceeding 500 million stock tank barrels) in Scottish maritime boundaries in the North Sea as well as the real oil price in constant 2001-USD. Panel B shows vote shares of SNP/Plaid Cymru in UK general elections and by-elections. The figure displays constituency results together with the total region-average vote share of the two parties.
One concern is that increased search efforts after the first oil discovery influenced the likelihood of future discoveries, but experts agree that the exact timing and size of oilfields are unpredictable. We thus exploit the timing of discoveries and the fact that few ‘giant’ oil fields are responsible for large shares of oil revenue as quasi-experimental variation in the regions’ relative resource wealth.
Another potential question of this type of analysis is whether other events that coincided with giant oil discoveries affected Scotland and Wales differently, for instance, the impact of Margaret Thatcher’s reform policies. We control for such events in various ways and make use of changes in the world oil price that affect the net present value of oil discoveries. Those changes provide quasi-random variation because the UK sells oil at the world market but, as a relatively small producer, it has little influence on the price.
Our statistical analysis shows that a single giant oil discovery increases the SNP’s vote share by about 2 percentage points. The effect is stronger when the world oil price is higher in absolute terms or relative to its long-term average, thus further supporting the hypothesis that potential revenues from oil are causally driving voters’ decisions. A back-of-the-envelope calculation suggests that oil discoveries contributed 12–16 percentage points to the initial rise of the SNP in the 1970s – about two-thirds of their vote gains during this decade.
General insights about democratic secessionism
What can we learn from these results to better understand the ongoing debates in the UK, Spain, and other countries? It is undisputed that cultural elements provide secessionist parties with potential leverage in developed democracies. To actually convince voters, however, parties need to move beyond that and provide compelling economic arguments for secession.
Those arguments are often based on the redistribution of fiscal revenues by a central government unit. This redistribution is blamed as unfair by secessionist parties and is combined with promises for better use in case of independence.
Our data suggest that publicly visible events like oil discoveries have a higher potential to move public opinion than actual oil revenues. This means that separatist economic claims do not necessarily need to be built on objective, verifiable benefits. For Catalonia and Scotland, the potential fiscal benefits of secession are still matters of much debate among experts than asserted by the respective secessionist parties. All this is reminiscent of Boris Johnson’s Brexit campaign, famously arguing that EU contributions can be better used for Britain’s own health system.
Finally, in light of the wide spread of separatist parties in Europe, it makes sense to look at the issue from the perspective of the EU. Support for European integration is often high in separatist regions. It would allow them to maintain the benefits of being part of a greater entity while overcoming and constraining to some extent the nation-states that they historically experienced problems with (Gehring 2020).
Still, with national governments acting as veto players at the EU level, the prospect of, for instance, allowing an independent Scotland fast-track access into the EU crucially depends on the strategic positions taken by national governments. As long as the actions of the European Commission continue to be influenced and be dependent on nation-state interests (Gehring and Schneider 2018), separatist parties’ EU affinity seems for now doomed to remain unrequited.
Alesina, A, and E Spolaore (1997), “On the number and size of nations”, The Quarterly Journal of Economics 112: 1027–56.
Collier, P, and A Hoeffler (2004), “Greed and grievance in civil war”, Oxford Economic Papers 56(4): 563–95.
Dehdari, S, and K Gehring (2019), “The origins of common identity: Evidence from Alsace-Lorraine”, CESifo Working Paper 7949.
Gehring, K, and S A Schneider (2018), “Towards the greater good? EU commissioners’ nationality and budget allocation in the European Union”, American Economic Journal: Economic Policy 10(1): 214–39.
Gehring, K, and S A Schneider (2020), “Regional resources and Democratic secessionism”, Journal of Public Economics 181.
Gehring, K (2020), “Overcoming history through exit or integration: Deep-rooted sources of EU support and Euroscepticism”, CESifo Working Paper 7831.
Morelli, M, and D Rohner (2015), “Resource concentration and civil wars”, Journal of Development Economics 117: 32–47.
MacKay, D I, and G A Mackay (1975), The political economy of North Sea oil, London: Robertson.
1 See https://knowledge.wharton.upenn.edu/article/secession-answer-case-catalonia-flanders-scotland/