The COVID-19 outbreak and the resulting disruptions in supply chains of some manufacturing and medical products have renewed the debate on costs and benefits of globalisation and, particularly, on risks associated with international fragmentation of production in global value chains (GVCs). While GVCs helped addressing supply shortages in several cases already during the early stages of the COVID-19 pandemic, much of the policy debate has concentrated on whether the gains from expanding international specialisation in GVCs are worth the associated risks of transmission of shocks and even whether governments should use policy tools to ‘re-localise’ GVCs. But re-localising may also mean less diversification and thereby limit the scope for cushioning shocks. This paper builds on on-going OECD analysis and aims at providing empirical evidence to inform and guide discussion on these questions. First, it reviews briefly the key issues and lessons learnt from the past, and identifies the main features of world trade and GVC participation that influence exposures to risks in supply chains. Subsequently, it presents key results of a set of economic model simulations conducted using the OECD’s computable general equilibrium (CGE) trade model METRO to shed light on the consequences of a stylised re-localisation policy scenario. In this scenario, countries are less exposed to foreign shocks, but they are also less efficient and less able to cushion shocks through trade. Quantitatively, the latter effect tends to dominate: re-localising GVCs would make the economy in most countries both less efficient and less stable. The economic case for policy-induced reshoring of GVCs is therefore weak. There is nevertheless scope for governments to join efforts with businesses to improve risk preparedness.