Common Fiscal Capacity Is Needed to Strengthen Risk Sharing

This asymmetric effect of the shock on economic activities becomes an asymmetric impact on countries depending on their sectoral composition.


Forum
regions specialised in services such as tourism particularly hard. Figure 1 shows a standard measure of employment specialisation across regions of the European Union. It reveals that regions with greater specialisation in retail trade are more affected by the pandemic. The impact of the shock is not only larger in regions specialised in contactintensive services but, most likely, will have a more persistent effect. There are two reasons for this: fi rst, substitution of labour for capital (robotisation) is much more diffi cult in non-routine jobs. Second, keeping safe physical distance implies that the very size of plants, shops, restaurants, etc. is a binding capacity constraint during the unlock phase. That is, restarting contact-intensive services in a way that complies with safety conditions requires a signifi cant amount of investment in the sectors that have been hardest hit by  Moreover, the shift in consumer expenditures from faceto-face services to online services (or goods) implies that there are sectors that are actually benefi tting from this crisis. In particular, the digital sector, dominated by international giants like Netfl ix, Google and Apple is making substantial profi ts in, almost literally, captive markets.

Risk sharing is needed to deepen economic integration
COVID-19 is an example of a shock that hits households, fi rms and sectors asymmetrically, within and across countries. This is particularly important in the European Union since all measures taken to boost the Single Market lead to exploiting comparative advantages across regions and, thus, sectoral specialisation; especially in the eurozone, as Mongelli et al. (2016) show. This specialisation raises the exposure of any regional economy to some risks and, therefore, increases the variance of its GDP. It could be argued that the COVID-19 shock is an extremely unlikely event, but there are many studies that point out that differences in sectoral composition are in fact an important factor in explaining the size of the business cycle fl uctuations in the regions as well as the asymmetries and fi nancial imbalances within the eurozone (see for instance, Imbs, 2004;Corsetti et al., 2008;Atalay, 2017). Kalemli-Ozcan et al. (2003) provide evidence that risk sharing and industrial specialisation are positively correlated using data that combines international and intraregional information. The message is that countries (or regions) that can shield consumption against production risk are better equipped to exploit the gains from specialisation much more and advance further in economic integration. The channels for risk sharing, aside from fi scal transfers, are banking and capital markets integration. Martinez et al. (2019) fi nd that a banking union is effi cient in sharing domestic demand shocks, while a capital market union is key to sharing supply shocks. That is, integration should go hand in hand on both fronts.

An incomplete risk-sharing architecture within the EMU
The question that arises is how countries in the EU, especially those in the Economic and Monetary Union (EMU), share the risks that arise from the industrial specialisation brought by the Single Market. Hoffman et al. (2019a) fi nd for the EMU area that, after the adoption of the euro, bank-

Figure 1 Employment specialisation across European regions, 2016
Note: The share of the total number of persons employed in each NUTS 2 region is computed for the six activities: a similar calculation is made for the whole of the EU28; the most specialised activity is computed by taking the regional shares and subtracting the EU28 shares; the map shows for each region the activity whose employment share exceeded the EU28 average by the largest margin (as measured in percentage point terms). Norway and Switzerland: national data. Germany, Greece, Spain, France, Cyprus, the Netherlands, Poland and Romania: provisional. Slovakia: estimates.
Agriculture, forestry and fishing Industry Construction Wholesale and retail trade; transport; accommodation and food services; information and communication Financial and insurance; real estate; professional, scientific and technical; administrative and support service activities Public administration (defence; social security; education; health and social work); arts, entertainment and recreation; others Data not available

Forum
ing integration grew signifi cantly through wholesale funding and the interbank market, which are highly procyclical (see Admati and Hellwig, 2014), instead of cross-border banking integration. This implies that risk sharing actually plummets during economic downturns, as we have seen during the European debt crisis in 2010-2012. As a matter of fact, sharing the same currency without advancing further in the capital market and cross-border banking integration leaves countries more exposed to asymmetric shocks and fi nancially fragile (Jaccard and Smets, 2020), especially if those countries do not share the same banking supervision. There have been advances in banking integration (see Bénassy-Quéré et al., 2018, for a discussion on this matter) and common banking supervision, but capital market integration is still lagging. The main obstacle to the latter is that the EMU lacks a safe asset. For capital markets to become deeper, abundant safe assets are needed to provide collateral; however, as we saw during the European debt crisis, sovereign bonds of countries facing fi scal tensions lose their safe asset status and, therefore, their value as collateral (Reis, 2019).
This problem is further amplifi ed by the so-called deadly embrace between sovereign debt and bank debt (Farhi and Tirole, 2018). The main casualty of the deadly embrace is credit to small fi rms, which cannot grow, hampering competition, as shown by Hoffman et al. (2019b). Thus, the implication is that, in the absence of a common safe asset, it is very diffi cult to break the deadly embrace and advance in cross-border banking and capital market integration, which are crucial to risk sharing.
This incomplete risk-sharing architecture is also being threatened by national responses to the pandemic. The same logic that applies to coordinating across regions within countries follows across countries in the EU, especially if we want to preserve the Schengen space. In the presence of externalities, uncoordinated actions lead to ineffi cient outcomes. The fact that some countries have more fi scal capacity than others implies that some countries can give more state aid to fi rms and sectors than other countries. This industrial policy at the member state level distorts competition and harms the Single Market, as Motta and Peitz (2020) argue. One could object that the asymmetric fi scal capacity is the result of differences in fi scal discipline, which is true. Nevertheless, we should bear in mind that tax revenues have a different elasticity across countries and, most likely, industry specialisation and the fi rm size distribution affect that elasticity. The estimates of Koester and Priesmeier (2017), and Mourre and Princen (2019) go in that direction.
The need for a safe common asset leads us, inevitably, to discussions of some sort of eurobonds. Mutualisation of debt among sovereign countries is not a good mechanism because it is plagued by many frictions, particularly limited commitment. But inaction is not an option because the current faulty architecture creates many economic and political tensions that endanger the European project. Additionally, faulty design can have unintended consequences. For instance, during the European debt crisis and the ensuing fl ight to safety, the return to German bonds reached historically negative levels. The question that arises is whether this was due to the fact that it was perceived as the only safe asset in euros. Finally, if we all agree that in order to make progress in banking and capital market integration, we need a safe asset, then we need to discuss common fi scal capacity as also argued by Pisani-Ferry et al. (2013). A safe asset will be valued if it is backed by tax revenues.

A common fi scal instrument conciliates risk sharing and fi scal discipline
The current coronavirus crisis has shown that uncoordinated policy responses lead not only to ineffi cient outcomes, but also to political tensions that give rise to anti-euro populism that may threaten the European project. A very stark example is the lack of coordination in health programmes and the protectionist reactions at the beginning of the pandemic. The disruption of global supply chains provoked by the virus has changed our understanding of the meaning of 'strategic industries' and there are calls to reduce country specialisations in a particular sector as a way to shield the economy against some shocks; that is, to reduce regional exposure to risk instead of sharing it. The underlying logic of this argument is that EU risk sharing mechanisms do not work well when needed so that it is better to smooth income by diversifying sectoral activities within countries. Thus, there is a serious protectionist threat that we have to deactivate.
The European Union is facing two big challenges: COV-ID-19 and climate change. They are both negative externalities; that is, both challenges call for coordinated action. We need to deepen risk sharing by means of common fi scal capacity and, at the same time, to coordinate policy against COVID-19 and the recovery phase of our economies while acting against climate change. But in doing so, we do not want to harm incentives: we do not want an enhanced EU fi scal capacity to backfi re and result in the reduced fi scal discipline of EU member states. Thus, the simplest solution is to create a common fi scal instrument to fi nance a common policy that should be discussed and designed in the common institutions. The EU already has the instruments to channel targeted policies across regions: the European Social Fund, the European Regional Development Fund, Horizon 2020, etc., and the European

Forum
Semester as the coordination net. This measure would complement the Stability and Growth Pact and the European Stability Mechanism.
The fi rst candidate for the common instrument is taxes directed at fi rms. There are three reasons, at least, why fi rm taxation should be done at the European level. First, the fact that fi rms can move headquarters and production plants easily across countries creates a problem for fi scal competition. Second, coordinated industrial policy can be undone by taxation at the national level. And third, the very size of many companies may give them strong infl uence at the national level but much less infl uence at the EU level.

EU common taxes to boost the Single Market
The fi rst step would be creating a common digital service tax. The project "Fair Taxation of the Digital Economy" (European Commission, 2018) includes two proposals: a reform of corporate tax rules so that fi rm profi ts are taxed where they accrue and an interim tax that covers digital activities currently untaxed in the EU. The important thing about this initiative is that it is targeting income that is not currently taxed. That is, it is aiming to solve a problem of horizontal justice as well as effi ciency, since not taxing activities distorts competition. As discussed above, the COVID-19 crisis has shifted household expenditures from face-to-face services to digital services. This huge shock has been positive for digital fi rms. The European Commission estimates that a 3% tax on digital services would generate €5 billion. To put this amount in perspective, notice that if the Commission raises €500 billion issuing bonds with a real return of 1%, the annual service would be exactly those €5 billion.
The second step would be to advance the Common Consolidated Corporate Tax Base (European Commission, 2016). Cross-border fi rms (and our goal is that EU fi rms grow enough to operate across borders) face 28 tax codes and have many opportunities for shifting profits and using loopholes. This fragmentation is a barrier to the Single Market, especially for small and mediumsized fi rms that want to grow as it is also an ineffi cient way to tax fi rms. According to the Commission (2016), 70% of fi rms' profi t is shifted across EU member states solely for tax purposes. This tax competition leads to a race to the bottom that is ineffi cient for two reasons. First, because it puts the burden of taxation on human capital accumulation. Second, without tax revenues we cannot build common public goods which are essential to fi ght the pandemic and climate change. According to the Commission estimates, EU businesses could cut their compliance costs by 2.5% under the common base and even more with the full consolidation of the tax base. Corporate taxes collected 2.7% of GDP across the EU in 2017 (see European Commission 2019). Collecting this revenue could be done in a much more effi cient way. A bolder step would be to phase out direct country contributions to the EU budget that are subject to much bargaining and negotiation and replace them gradually with revenues from the common corporate tax.
Additionally, to boost a green recovery, the EU Emissions Trading System needs to be clarifi ed and, perhaps, allowances should be restricted. The auction system should be revised taking into account the dynamic responses of fi rms. We should also consider introducing a carbon tax, which levied upstream on the fuel itself when it is extracted or imported, simplifi es its administration and the management of carbon leakages. These two ways of taxing fi rms should be coordinated within the common corporate tax base.

The political economy of risk sharing
Finally, we should take into account the fact that the gains of improving risk sharing across EU members and the costs of reforming current rules may occur at different horizons. Economies are dynamic and change over time.
Not only that, the fact that countries have heterogeneous sectors and different income and wealth distributions implies that there are always winners and losers within countries. Thus, we should study further the gains of improving the mechanisms of risk sharing across countries, taking into account the underlying heterogeneity within EU country member states. In particular, the unanimity rule in the Council of the EU is a distortion to risk sharing since many side payments, transfers or budget rebates are agreed to circumvent veto power of member states regardless of effi ciency criteria. We should replace the unanimity rule with a qualifi ed majority rule. In this way we can design better policies so the net gains of risk sharing are felt by all citizens within the European Union.