Raphaël Lafrogne-Joussier, Andrei Levchenko, Julien Martin, Isabelle Méjean April 24, 2022
Editor’s note: This column is part of Vox’s debate on the economic consequences of war.
Imports from Russia are largely made up of energy inputs, including oil, coal and natural gas. The centrality of these inputs in production networks implies that shocks affecting the price of energy have the potential to propagate downstream, leading to a significant amplification of the shock. However, recent estimates, drawn from calibrated multi-sector, multi-country models with input-output linkages (Bachmann et al. 2022, Baqaee et al. 2022), suggest that the effect of an oil and gas import ban would generate a relatively limited contraction in GDP.1 Even in a country like Germany, cutting Russian energy imports – which account for 30% of German energy consumption – would lead to a 0.5 to 3% drop in GDP, a significant but manageable economic cost.
In sectoral models such as those used in Bachmann et al. (2022) or Baqaee et al. (2022), such a relatively small effect results from a non-zero elasticity of substitution between firm inputs. If Russian oil and gas were perfect complements to other inputs (zero elasticity), GDP would fall along with energy imports. Some of the companies in the sectors dependent on Russian energy are supposed to be able to turn to other suppliers and the same goes for companies in the downstream sectors, which have to deal with the reduction of production by their suppliers. In particular, the open economy structure of the model implies that some of the inputs that can no longer be produced locally due to energy rationing can be replaced by foreign goods.2
The assumption that there are opportunities for substitution in production networks may seem controversial, as they are generally viewed as rigid structures shaped by relationship-specific investments. Is it more appropriate to be rather conservative and assume a purely Leontief production structure? Answering this question is tricky in the absence of direct evidence on how technologies adapt to energy shocks.
Is Leontief technology at the enterprise level?
In a recent article (Lafrogne-Joussier et al. 2022), we address a related question by using the early stage of the Covid crisis as a quasi-natural experience. In this empirical study, we use monthly panel data on French companies and investigate their sales dynamics in the first half of 2020. Our strategy exploits companies’ early exposure to lockdown-induced supply chain disruptions in China in January 2020. By comparing companies that have been exposed to the productivity slowdown in China across their value chain with comparable companies that have not, one can quantify the magnitude of the propagation of the shocks, to the origin of formalized amplification in production network models. In March 2020, when the virus was just beginning to spread in Europe and France, companies exposed to supply chain disruptions from China were already showing export sales 7% lower than those from China. their unexposed counterparts.
The exercise also makes it possible to examine the heterogeneous adjustments of firms exposed to input disturbances. A first finding is that exposed companies holding inventories have better cushioned the shock. Whether they are managed at company level, as in our example, or by the public authorities, strategic stocks (in particular of gas and oil) appear to be essential in helping exposed companies to mitigate the shock. A second, more surprising result is that firms that relied the most on Chinese inputs absorbed some of the shock by diversifying their supply chain after the early lockdown. Among the processing group, firms that had an undiversified supply chain have a significantly higher likelihood of starting to import their inputs from elsewhere right after the Chinese lockdown-induced disruption, in February and March 2020 (Figure 1 ). While such evidence does not directly address the question of substitution possibilities for Russian gas, it does support the idea that, even in the very short term, companies facing significant disruptions in their input purchases are adapting. .
Figure 1 Impact of early foreclosure in China on the number of foreign partners of exposed companies.
Source: Lafrogne et al. (2022).
Remarks: The figure shows the result of an event study that compares companies exposed to Chinese inputs before the lockdown in China (“treated” companies) and companies that were not (“control” companies). The processing group is further divided into “diversified” companies that were linked to at least one other sourcing country for inputs from China and “non-diversified” companies that were solely dependent on China before the shock. The estimated equation explains the number of source countries, before and after January 2020, in the group of treated companies compared to the control companies, using a Poisson estimator. The difference is normalized to zero in January 2020.
Distributive effects of shock
Another margin of adjustment, which the textbook industry models do not incorporate directly, is substitution within an industry between firms. Since firms in the same industry produce output that is likely to be more highly substitutable than inputs within the firm, the heterogeneity of Russian gas users provides another shock-mitigating mechanism (di Giovanni et al. 2020). As discussed in this article, the heterogeneity of exposure to a foreign shock has important global consequences on the global impact of the shock. The heterogeneity of exposure to Russia on the import side is illustrated in Figure 2. Out of 150,000 French importers, less than 2,500 imported directly from Russia in 2019. However, these companies are significantly larger than average and their total imports represent one third of the total imports of France. If the companies exposed are large and linked to other domestic producers, their sensitivity to the shock has global consequences. But heterogeneity also has distributional consequences: unexposed firms gain market share over exposed firms. To account for these substitution opportunities, the analysis by di Giovanni et al. (2020) maps firm-level data for France with industry-level input-output data used in Bachmann et al. (2022) or Baqaee et al. (2022).
Figure 2 Exposure of French companies to Russian imports
Source: French customs data for 2019.
Remarks: The figure shows the number (left panel) and share in overall imports (right panel) of firms that i) import from Russia (dark gray bar), ii) import one of their inputs only from Russia (gray bars clear) and iii) import one of their main inputs only from Russia (blue bars). In the third case, the statistics are based on the subsample of a company’s imports that represent at least 1% of the company’s overall imports in 2019.
Figure 3 illustrates how the heterogeneity of exposure and substitution opportunities affects the response of French firms to a 10% drop in Russian productivity. According to our basic calibration, the aggregate impact of such a shock is a 0.9% decline in France’s real GDP (the red line in Chart 3). The blue circles show the average firm-level responses by firm size. While companies in the top two percentiles of the size distribution experience a large adjustment of 4%, some companies in the lower percentiles grow as they gain market share from the most exposed companies. These substitution opportunities are not taken into account in classical models with input-output links, but they could be important in the context of a possible ban on Russian gas if there is heterogeneity between firms in a sector. in their dependence on Russian gas.3
picture 3 Heterogeneity of companies’ response to a 10% drop in productivity in Russia
Source: calculation by the authors using the model of di Giovanni et al. (2020).
Remarks: The figure shows the average elasticity of companies’ real value added to a simulated 10% decline in the overall productivity of the Russian economy. The average elasticities are calculated for 50 slices of individual companies, grouped according to their size (value added).
Existing evidence from detailed firm-level data therefore supports the view that foreign shocks do indeed diffuse through production networks. Despite the rigidity of modern production networks, some companies adjust their technology, even in the very short term, when faced with a break in their value chain. Moreover, heterogeneous exposure to the shock has distributional consequences: the least exposed companies gain market share over the most exposed. The assumption of some substitution between inputs in production network models is consistent with this micro-level evidence. But what the discussion also shows is that a ban on Russian imports will have very heterogeneous consequences. Some well-known companies and iconic products will be heavily affected by the sanctions. Beyond GDP figures, huge but concentrated losses can have a stronger impact on public opinion than small diffuse losses.
Bachmann, R, D Baqaee, C Bayer, M Kuhn, A Loschel, B Moll, A Peichl, K Pittel and M Schularick (2022), “What if Germany was cut off from Russian energy? », VoxEU.org, 25 March .
Baqaee, D, C Landais, P Martin and B Moll (2022), “The economic consequences of stopping energy imports from Russia”, French Council of Economic Advisors.
di Giovanni, J, AA Levchenko and I Mejean (2020), “Foreign Shocks as Granular Fluctuations”, CEPR Discussion Paper 15458.
Lafrogne-Joussier, R, J Martin and I Mejean (2022), “Supply chain disruptions and mitigation strategies”, VoxEU.org, 5 February.
1 Importantly, this result is retrieved from a model in which the only source of such a contraction is attributable to the propagation of the shock in production networks. Other negative consequences such as the adverse effect of the wealth shock induced by the rise in energy prices are assumed to be neutralized.
2 Of course, the substitution of domestic production by foreign inputs has other consequences. Notably, higher imports imply more production abroad, and therefore higher energy demand.
While this substitution between firms within a sector should help to absorb the shock, it is difficult to quantify the magnitude of this damping mechanism because it varies according to the elasticity of substitution between the output of firms and the extent to which factors can be reallocated between firms. In the benchmark calibration, the elasticity of substitution is set to 3 and labor is assumed to reallocate between firms without friction.