Product margin adjustment to crises

Karen-Helene Ulltveit-Moe, Andreas Moxnes

01 October 2010



The global crisis hit world trade hard – and fast. According to the WTO (2009), in barely a year between early 2008 and early 2009, world trade in manufactures fell by nearly a third. The great trade collapse, as it came to be called (see Baldwin 2009), has opened up a whole new debate over its causes and consequences.

In a Vox column this spring, Behrens et al. (2010) presented empirical evidence suggesting that the collapse was driven by falling quantities, and to some extent, by falling prices. In the short run, almost all trading firms remained active. There was hardly any change in the average number of countries they traded with, or in the average number of products shipped to or sourced from each country. In other words, there were few signs of output-side adjustments in the short run. In another Vox column, Haddad et al. (2010) use data from several countries to show that the great trade collapse was indeed more concentrated along the intensive margin – the reduction in the value of goods already being traded.

As these studies show, restructuring implies adjustments on firms’ input- or output-side. In our recent analysis based on Norwegian manufacturing (Moxnes and Ulltveit-Moe 2010), we hope to shed some light on the prospects for output-side adjustment following a crisis – in this case the 2001-2002 real-exchange-rate shock. We investigate the rate of firm-level product “churning” in the aftermath of a trade shock. Product portfolio churning is measured as the sum of firm-level adds and drops of exported products from one year to another, and therefore proxies one aspect of within-firm restructuring. For the group of continuous exporters that we consider, the great majority are multi-product firms. On average, each firm exported 14 products (defined as unique HS8 varieties). As an indication of the importance of firm-level restructuring, each of these firms was on average churning roughly 15 products from one year to the next (the median is 9).

The 2001-2002 real-exchange-rate shock in Norway was characterised by a 17% appreciation over a short period of time and a 9% decline in exports for that year. The advantage of analysing this particular episode is that it allows us to construct a counterfactual, namely what would have happened in the absence of the shock? At the level of the firm, some firms had a natural hedge against the shock as they were equally exposed to international markets on the input (imports) and output (exports) side. We use this heterogeneity in exposure to construct two groups of firms: a control group which was ex-ante not exposed to the shock and a treatment group which was ex-ante exposed. We then employ a simple difference-in-difference strategy and compare the effect of the shock on firm-level product churning for the treatment and control group.

Our findings suggest that the real exchange rate shock induced firms to slow down or postpone the introduction of new products to the market, which translated into fewer additions to firms' product range and less firm-level product churning. Moreover, churning was the firms’ main margin of adjustment, not the net number of products exported. In that sense, we only find weak support for the core-competencies hypothesis put forward by Bernard et al. (2006).

A cause for cheer? Or a cause for concern?

The lack of restructuring of firms’ product mix in the aftermath of the crisis sounds like bad news for productivity. Or to put it more mildly, it is not likely that this type of output-side adjustment contributes to substantial productivity gains`.

Our evidence could potentially be bad news for European manufacturing. A sustainable recovery requires restructuring. Rigid European labour markets suggest that this will not happen through reallocation of labour. Hence, focusing on core competence and innovation that foster new products and product churning is the way ahead. Our analysis of a recent trade crisis in Norwegian manufacturing suggests that this will take time.

So what to do? We argue that there is a role for policy supporting research, innovation, and the development of new products, in order to ensure restructuring and growth throughout Europe.


Baldwin, Richard (ed.) (2009), The great trade collapse: Causes, consequences, and prospects, A Publication, 27 November.

Bernard, AB, SJ Redding and PK Schott (2006), "Multi-Product Firms and Trade Liberalization", NBER Working Paper 12782 (revised 2009)

Behrens, Kristian, Gregory Corcos, and Giordano Mion (2010), “Trade collapse or trade crisis?”,, 21 March.

Haddad, Mona, Ann Harrison, and Catherine Hausman (2010) “Prices vs quantities in the great trade collapse”,, 27 August.

Ekholm, K., Moxnes, A. and K. H. Ulltveit-Moe (2008): “Manufacturing restructuring and the role of real exchange rate shocks: a firm level analysis”, CEPR Discussion Paper 6904,

Moxnes, A. and K. H. Ulltveit-Moe (2010): “Product adjustments: a firm level analysis of the impact of a real exchange rate shock.”, CEPR Discussion Paper 7923



Topics:  Global crisis International trade

Tags:  international trade, global crisis, great trade collapse

Professor of International Economics, University of Oslo; and CEPR Research Fellow

Associate Professor of Economics, University of Oslo