Larger firms respond more strongly to macro shocks, and this is important

Jean-Charles Bricongne, Juan Carluccio, Lionel Fontagne, Guillaume Gaullier, Sebastian Stumpner 27 July 2022

We know from the seminal contribution of Gabaix (2011) that changes in the performance of a few very large firms matter for aggregate outcomes in a granular economy. The ‘micro to macro’ approach, linking micro behavior to macro outcomes, has substantially improved our understanding of macro aggregates such as business cycles, comparative advantage (Gaubert and Itskhoky 2020) and the international transmission of shocks (Di Giovanni et al. 2012).

Because changes in the performance of these large firms are important for macroeconomics, understanding their roots is paramount. Why do large firms operate differently than small ones? While the literature has focused on the role of idiosyncratic shocks (Kramarz et al. 2019), a complementary perspective reveals that large firms have differential responses to common shocks affecting all firms. This approach posits that macro shocks lead to heterogeneous responses, especially by the largest firms, which in turn determine the macro response to the initial shock – ie from macro to micro to macro. In a recent paper (Bricongne et al. 2022), we analyze the contribution of the largest exporters to total export fluctuations over a longer period spanning 1993 to 2020. We rely on a universe of detailed firm-level export data collected by the French. Customs offices contain export values ​​by country of destination in finely defined commodity codes and, importantly, are available on a monthly frequency.

In Figure 1, we decompose aggregate export growth (in quarterly frequency for readability) into an unweighted average of firm export growth rates and a granular residual. The latter captures the equivalence between firm size and firm growth. If the macroshock response is uncorrelated with firm size, the granularity residual will be zero. The granular residual is not zero, and furthermore, it explains a large part of aggregate export fluctuations: 42% of the variance in aggregate export growth. Furthermore, the correlation coefficient between the unweighted mean firm growth and graininess residual is close to 0.5. This implies that large exporters tend to do worse than average firms in recessions and better than average in booms.

Figure 1 Average firm export growth and granular residuals

Disagree: The mid-point growth rate of aggregate quarterly French exports is decomposed into the unadjusted average growth rate across exporters (blue line) and exporter size and unweighted growth rate (granular residuals, red line).

Large exporters have driven export declines in the global crisis and pandemic

The overreaction of large firms to macro shocks is large and evident in the case of the two largest macro global shocks of the past decade, where the decline in French exports was of similar magnitude (-17.4% for 2009/2008 and -16.3% for 2020/2019). Not only are the two export declines almost entirely explained by intensive margins (firms that continue to export), but they are also caused by the largest exporters, whose export growth rates were significantly lower than the average exporter.

We illustrate this in Figure 2, where we plot weighted average year-over-year mid-point growth rates by non-overlapping size bins of exporters. Size bins are defined using the pre-crisis exporter size distribution (2019 for Covid and 2008 for the global crisis). Growth rates in terms of sectoral and geographic profiles of firm-level exports were cleaned of composition effects and thus calculated as growth in exports within finely defined markets. The top 0.1% of exporters (about 100 firms out of 100,000) are represented by the red line.

The message is clear: growth in top exporters has slowed significantly compared to average exporters, controlling for compositional effects in terms of sector and destination. This pattern holds in both crises. Interestingly, in both events, the largest exporters also experienced a slower recovery than the bottom 90%.

Figure 2 Growth rate of exports during the Covid crisis (left) and global crisis (right), by Akar Bin

Disagree: 12-month weighted average mid-point growth rate by decile of exporter size distribution. Exporter size bins are defined using the pre-crisis distribution of export size distributions (total firm-level exports in 2019 for the case of Covid and total firm-level exports in 2008 for the global crisis).

We zoom in on the April and May 2020 export declines in Figure 3. Given the huge concentration of exports, we choose particularly fine bins at the top of the distribution. For example, the top 1% (about 1,000 firms) account for more than 70% of total exports. The black bars show the share of total exports for April and May 2019 attributable to each size bin.

We then compare the pre-crisis export share of each bin Contributing to the decline in overall exports Between April and May 2019 and April and May 2020, the change in total exports of a bin is measured by dividing the change in total exports. If all firms grow at the same rate, each bin’s contribution will be equal to its pre-crisis share. The figure shows that a small group of ‘superstar’ exporters disproportionately explain the export slowdown. The top 0.1% of exporters contributed 57% of the decline in overall exports, compared to their pre-crisis share of just 41%. Among the top 0.1%, the ten largest exporters alone accounted for nearly one-third of the export decline, when they exported 19% of total pre-crisis value. The message is the same as in Figure 1. The negative relationship between pre-crisis size and export coordination with the crisis is also present in the set of 1,000 largest exporters.

Figure 3 2019 Covid-19 Export Share and Contribution to Trade Growth 2019-2020 by Size Bin

Disagree: Pre-crisis export share and contribution to overall export decline between April and May 2019 and April and May 2020. Exporter-size bins are constructed using 2019 export prices by firms.

The 2020 decline in French exports was driven by demand shocks; Global value chain disruption has played a lesser role

The Covid-19 pandemic provides us with an excellent laboratory to study the role of heterogeneous responses to collective shocks. The shock was sudden and external. Although sanitary measures were imposed in most French trading partners, their time variations suggest that we can exploit, thanks to monthly frequency data, to measure both supply and demand shocks.

Larger firms are indeed more likely to be involved in complex global value chains (GVCs) (Antras 2020) and more likely to experience supply disruptions caused by systemic shocks (Baldwin and Freeman 2022). Our aim is to understand whether the greater GVC exposure of top exporters can explain their stronger response to shocks, not whether GVCs themselves matter. We supplement the export data with firm-level import and sales information and measure each exporter’s GVC exposure using the Imported Intermediate Inputs to Sales (IIS ratio) and supply shock exposure using information on lockdowns in import source countries. . We develop a flexible regression framework that relates the growth rate in each market (defined as a product-destination pair) to size bin dummies. The data reveal that adding the GVC measure to our regression does not affect the magnitude and significance of the exporter size-bin dummies. In other words, the overreaction of large exporters was not due to their deep engagement in GVCs.

In contrast, we find strong evidence of a demand channel that is not driven by exporters’ sector or destination composition. Instead, we hypothesize a greater resilience of large firms to destination-country lockdowns. Specifically, we regress each month on the firm-product-country-month level midpoint growth rate on the Oxford Stringency Index (Hale et al. 2021). The identification exploits the variation in export growth of the same firm across destinations with different degrees of lockdown, fully controlling for product-level shocks. The regression fully controls for firm-level supply shocks, originating in France and abroad, including firm*month fixed effects. The results are shown in Figure 4. On average, the median growth rate decreased by 0.6 points from full to no lockdown. However, the effect is strongly heterogeneous, almost double for firms in the top 0.1% (1.0) and the bottom 99.99% (below 0.5).

Figure 4 Destination lockdown effect by size bin

Disagree: Lockdown stringency is interacted with a set of six complementary size dummies, including a regression with firm-month, product-month and destination fixed effects. The dependent variable is the midpoint growth rate of exports by firm, product and destination country in a given month. We plot point estimates and 1% confidence intervals.

Identifying the role of large firms for macroeconomic aggregation is a vibrant and critically important area of ​​research. This has various implications for economic policy-making (see, for example, a plea for Russian energy imports, Lafrogne-Jussier et al. 2022). Our results show that the response of aggregate exports to large macroeconomic shocks is largely driven by the larger weight of large firms in the economy and their greater sensitivity to these shocks. Export champions’ extremely high contribution to commercial success can thus become a vulnerability in the event of a sudden downturn in the business cycle.


Antras, P (2020), “Conceptual Aspects of Global Value Chains”, World Bank Policy Research Working Paper 9114.

Baldwin, R and R Freeman (2022), “Global Supply Chain Risk and Resilience”,, 6 April.

Bricongne JC, J Carluccio, L Fontagne, G Gaullier and S Stumpner (2022), “From macro to micro: large exporters coping with common shocks”, Bank of France Working Paper 881.

Di Giovanni, J, A Levchenko and I Méjean (2012), “The Role of Firms in Aggregate Fluctuations”,, 16 November.

Di Giovanni, J, A Levchenko and I Méjean, (2020), “Foreign Shocks as Granular Fluctuations”, Bureau of Economic Research Working Paper 28123.

Gabaix, X (2011), “The Granular Origin of Aggregate Fluctuations”, Econometrica 79(3): 733-772.

Gaubert C and O Itskhoki (2020), “Comparative Advantage of Superstar Firms and Countries”,, 14 August.

Hale, T, N Angrist, R Goldszmitt, B Kira, A Petherick, T Phillips, S Webster, E Cameron-Blake, L Hallas, S Majumdar and H Tatlow (2021), “A Global Panel Database of Epidemic Policy (Oxford Covid) -19 Official Feedback Tracker)”, Nature is human behavior 5: 529-538.

Kramarz, F, J Martin and I Méjean (2019), “Idiosyncratic risks and the volatility of trade”,, 11 December.

Lafrogne-Joussier, R, A Levchenko, J Martin and I Méjean (2022), “Beyond macro: firm-level effects of Russian power cuts”,, 24 April.

Leave a Reply

Your email address will not be published.