On the correction of Eurozone external imbalances and the pitfalls of bilateral imbalance measures

Filippo di Mauro, Arne Nagengast, Robert Stehrer

29 January 2016



The worst phase of the Eurozone (EZ) crisis is over and the countries concerned have altogether been able to substantially readjust their external imbalances with a combination of export creation and demand contraction (European Commission 2015). Where do we go from here? Three questions naturally emerge:

  • Is there a need for further adjustment?
  • If so, what needs to be adjusted?
  • Should the objective be to improve bilateral current account balances, and what combination of tools are actually available to EZ policymakers in order to do so?

Out of the three, this column takes up the third question and intends to challenge common wisdom. Let’s suppose an EZ deficit country has to correct its total current account balance. The typical policy suggestion would be twofold. Within the Eurozone, policies would need to aim at lowering relative export prices in the common currency via readjustments of wages and prices in order to improve the trade balance with EZ partners. This price/cost readjustment – eased by an overall depreciation of the euro – would also bring about improvements in the trade balance with the rest of the world, thus fostering overall external rebalancing. Second, recommendations would include a call for moderating both public and private demand (typically to correct pre-crisis ‘excesses’) in order to support the adjustment of the trade balance both with other EZ countries and the rest of the world.1

This is the typical advice, which is in principle sound. When it comes to implementing it, however, a critical question comes to the fore: How much control do EZ countries actually have over their internal EZ trade balances? In other words, can we really expect that domestic policies aimed at dampening prices and costs, as well as curbing excessive demand, would effectively help correcting them? This column argues that this is not necessarily the case, at least not to the same extent for all EZ countries. Why? Because of the emergence of global value chains (GVCs), into which EZ economies are tightly integrated. In such circumstances, trade imbalances are to a large extent an endogenous result of the international organisation of production at the firm level. The implication is that imbalances within the EZ tend to be increasingly less affected by the adjustment of relative prices and demand vis-à-vis EZ trade partners.

Assessing the impact of global value chains on trade imbalances

Global value chains can greatly distort the reported trade flows between countries (Koopman et al. 2014, Johnson and Noguera 2012) and can generally undermine the validity of standard competitiveness indicators, as argued in several reports by the ESCB’s Competitiveness Research Network CompNet (e.g. di Mauro and Ronchi 2015). The issue is that official trade statistics are reported in ‘gross’ terms, which measure the value of the physical flow of goods (and services) across borders. The prevalence of trade in intermediates implies that countries do not necessarily consume the goods that they import, but rather that they often use them as inputs in the production of other final and intermediate goods that may again be exported to other destinations. In order to correct for this, a number of papers have proposed alternative measures which extract the value added embedded in gross trade flows. Hence, value-added trade statistics capture the origin of the value added of final goods that are consumed in any given country.

Value-added trade flows can differ substantially from gross trade flows, and this is particularly the case in the Eurozone where cross-border production networks have dramatically increased in the last two decades (Amador and di Mauro 2015). Figure 1 shows that between 1995 and 2008, intra-EMU19 trade imbalances increased substantially as measured by the standard deviation of the bilateral gross trade balances between all EMU19 countries. Rather remarkably, however, over the same period there has been a growing divergence between the measure in gross terms with respect to the one in value added.

Figure 1. Standard deviation of intra-EMU19 bilateral trade imbalances

Source: Nagengast and Stehrer (2014).

Figure 2 provides details on the source of the difference between gross and value-added trade balances. The bulk – about a third, but constant over time – is due to value added generated abroad (foreign value added) and directly absorbed by one of the two direct trading partners. However, the most dynamic factor over the period has been the demand in third countries, which increased from about zero in the mid-90s to 20% in 2011, as a result of the expansion of intra-EZ production networks (Nagengast and Stehrer 2014). The importance of this third-country effect implies that a sizeable portion of gross bilateral trade balances can no longer be influenced directly by demand conditions in the two respective trading partners.

Figure 2. Variance decomposition of intra-EMU19 bilateral gross trade imbalances

Source: Nagengast and Stehrer (2014).

Since the total trade balance of any individual country is by definition the same when measured in gross or value-added terms, the above evidence suggests that, overall, intra-EZ trade imbalances were overstated when measured, as they usually are, in gross terms, while those with countries outside the EZ were correspondingly underestimated.

Let’s look at some actual numbers for individual countries. Table 1 details data on all EZ countries’ value-added trade balances with the Eurozone as a percentage of their gross trade balance with the Eurozone, using data from the WIOD between 2008 and 2011. This is the latest period available and a four year average has been used to decrease the volatility of the results. The same conclusions emerge when using other data sources (such as the OECD TiVA), although the ratios differ in magnitude which needs to be addressed in future efforts of creating multi-country input-output tables. The main takeaway messages are as follows:

  • For the vast majority of EZ countries (16 out of 19), trade balances measured in value added terms vis-á-vis the Eurozone as a whole were smaller (and therefore those with non-EZ countries larger) than the ones measured in gross terms.
  • Looking at deficit countries, the ratio was particularly low for France, Portugal, and Italy (in increasing order). This implies that the trade balance with the Eurozone in gross terms would be increasingly less amenable to the adjustment of domestic demand and relative prices of these countries with their EZ trade partners.

Table 1. Selected trade balance measures of individual EZ countries

Source: Authors’ calculations based on WIOD.

Policy implications

Does all of the above imply a call for policymakers to look at bilateral balances – and most notably at EZ imbalances – only when measured in value-added terms and thus base their policies exclusively on these statistics? Yes and no.

In support of that view are the following two points:

  • As explained above, bilateral value-added balances account and correct for third country demand effects, foreign value added, and double counting in trade flows (Koopman et al. 2014, Nagengast and Stehrer forthcoming).
  • Value-added flows also take into account that gross exports often contain a large share of domestic services value added, which is critical when assessing adjustment needs of relative prices (Bems and Johnson 2012, Patel et al. 2014).

However, at least two objections can be raised:

  • A value-added perspective abstracts from the actual physical flow of goods across countries and gives the impression that countries directly trade value added with each other.2

However, due to the trade in intermediates, a large part of value added crosses international borders multiple times before being absorbed in final demand (Nagengast and Stehrer forthcoming). This means that relative price changes can have complex effects. For example, a loss in competitiveness of a surplus country might drag down the exports of a deficit country, whose value added is indirectly being exported by the former (Bems and Johnson 2012).

  • Another important caveat is that while the overall results are broadly consistent across databases, there are a number of divergences regarding the precise level of bilateral gross and value-added trade balances.

These data issues will need to be addressed in future efforts to create multi-country input-output tables, which are the starting point for computing statistics in value-added terms.


In conclusion, the available data on value-added trade balances clearly show that assessing intra-EZ imbalances in gross terms is subject to serious drawbacks. In particular, it appears that – for most countries – a large part of intra-EZ imbalances is increasingly independent of the adjustment of domestic demand and relative prices vis-á-vis EZ trade partners. Focusing on bilateral value-added trade balances appears to get around some of the issues, but itself has some disadvantages, at least until our economic understanding of price competitiveness in global value chains and the underlying statistical data improve. At the current juncture, the best advice economists can give policymakers is what they have always told them in the past – focus on the aggregate trade balance and support the construction of reliable statistics in order to account for the role of global value chains as a major driver of trade and country competitiveness now and in the years to come.

Disclaimer: The views expressed here are those of the authors and do not necessarily represent those of the Deutsche Bundesbank or the European Central Bank.​​


Amador, J and F di Mauro (2015), The Age of Global Value Chains, A VoxEU.org eBook, CEPR Press.

Bems, R and R C Johnson (2012), “Value-Added Exchange Rates”, NBER Working Paper No. 18498, Cambridge, MA.

di Mauro, F and M Ronchi (2015), “Assessing European Competitiveness: The Contribution of CompNet Research”, ECB CompNet Report (June 2015).

European Commission (2015), “European Economic Forecast – Spring 2015”, European Economy 2/2015.

Johnson, R C and G Noguera (2012), “Accounting for Intermediates: Production Sharing and Trade in Value Added”, Journal of International Economics 86(2), pp. 224–236.

Koopman, R, Z Wang and S-J Wei (2014), “Tracing Value-Added and Double Counting in Gross Exports”, The American Economic Review 104(2), pp. 459–94.

Nagengast, A J and R Stehrer (forthcoming), “Accounting for the Differences Between Gross and Value Added Trade Balances”, The World Economy.

Nagengast, A J and R Stehrer (2014), “Collateral Imbalances in Intra-European Trade? Accounting for the Differences Between Gross and Value Added Trade Balances”, ECB Working Paper No. 1695, Frankfurt am Main.

Patel, N, Z Wang and S-J Wei (2014), “Global Value Chains and Effective Exchange Rates at the Country-Sector Level”,  NBER Working Paper No. 20236, Cambridge, MA.


1 The opposite policy mix would usually be prescribed for surplus countries.

2 When assessing the price competitiveness of countries, this view is only valid for a knife-edge case of identical elasticities of substitution in final demand and production (Bems and Johnson 2012).



Topics:  EU policies International trade

Tags:  trade imbalances, eurozone, value-added trade balances, global value chains

Senior Adviser in the Research Department, European Central Bank; Chairman, CompNet

Economist, Deutsche Bundesbank

Senior Economist and Deputy Scientific Director, Vienna Institute for International Economic Studies