The German economic recovery from the Great Recession has been robust. This has led to calls for Germany to play a bigger role in assisting other European economies, particularly those along the continent’s periphery (Posen 2011). This raises the question whether Germany can, in fact, play a locomotive role with autonomous impulses that are over and beyond the impulses generated from the broader trends in global trade.
How likely is this? The key to answering this question is that variations in German GDP are driven largely by external forces, more so than for any other advanced economy. This phenomenon has been clear in the recent crisis and recovery. The initial impulse to both Germany’s economic contraction during the crisis and its subsequent recovery came from global growth.
Thus, because Germany’s own domestic demand has been largely responsive to international cyclical conditions, Germany has not acted, in the short-term, as an independent source of growth. Put differently, global growth acceleration and deceleration have had simultaneous influences on Germany and the rest of Europe. Germany’s trading partners have received the global impulses directly and also via Germany, as stepped-up German exports have drawn in more imports.
Germany does generate some independent impulses within Europe, mainly to neighbours with strong trade links and to new EU member states through increased offshoring and other strengthening of economic ties (see Danninger 2008). In contrast, the limited trade links between Germany and the European crisis countries mean that any spillovers of German growth to those countries have also been limited.
Recent research on international spillovers has focused on how the US affects growth elsewhere in the world through its financial conditions or demand for other countries’ exports (see IMF 2007, Bayoumi and Swiston 2009). The studies generally find significant effects from the US economy on other countries and little feedback to the US. Spillovers from the Eurozone or Japan are found to be smaller than spillovers from the US. There is also evidence that US spillovers to other regions were unexpectedly high during the recent 2008-09 financial crisis. In particular, the impact of the US-based on a structural vector auto regression (SVAR) analysis and on a sample that includes the 2008-09 crisis period is almost twice as large as elasticities estimated on pre-crisis data (Swiston 2010). This suggests that special transmission channels – or more pronounced play of the same channels – are at play during crisis times that cause stronger cross-border spillovers. The potential for these spillovers to transform a business-cycle correction originating in the US into a crisis of global dimensions was illustrated by the synchronised global downturn of 2008-2009.
In a recent study (Poirson and Weber 2010), we have expanded the SVAR-based analysis to a wider set of 17 OECD countries and a more recent period that includes the crisis.1 Unlike earlier studies which treated the Eurozone as an aggregate, we incorporate interactions both within individual Eurozone countries and between Europe and other regions. We estimated the model using quarterly PPP-adjusted GDP for the full sample and a more recent sub-sample starting in 1993, and both including and excluding a 2008-09 crisis dummy. Following Bayoumi and Swiston 2009, identification is obtained via 48 different orderings, which give larger countries a proportionately greater probability of being the “growth leader”. An average impulse response across the different orderings is used to measure the spillover effects. Since the model imposes no structure on the data, the coefficient estimates reflect all relevant channels of transmission of shocks, including both financial and trade channels.
Our main finding is that outward spillovers from Germany’s growth to other countries have been low and have remained modest in recent years. In contrast, spillovers, especially from the US, followed by the UK as a distant second, have been larger and have increased over time, even after controlling for the effect of outsized spillovers during the crisis. Based on full sample estimates including a crisis dummy, the effect of a 1% growth shock in Germany measured by the peak cumulative response of other countries is about 0.1%, the lowest of all the large countries. Moreover, spillovers from Germany have decreased in recent years, halving to only 0.05%. Japan’s impact on the rest of the world has similarly decreased.
Figure 1. Outward spillovers from large systemic countries/regions 1/
Source: OECD, WEO, and IMF staff calculations. Notes: Sample includes 17 OECD countries, and all specifications include dummy variables for the 1979 and 1990 oil shocks and the 2008-09 crisis. 1/ GDP-weighted average response of other countries. 2/ GDP-weighted average of outward spillovers from non-German euro area countries (Austria, Belgium, Finland, France, Greece, Ireland, Italy, the Netherlands, Portugal, and Spain).
The relatively small size of German spillovers is a surprise, given its large economy. Indeed, Germany falls significantly below the regression line – generating much smaller spillovers than its size may suggest. This difference is related to the other determinant of spillovers. The analysis suggests that countries reliant on external rather than domestic sources of growth generate smaller spillovers. This is the case with Germany, whose growth is powered by global growth, which implies that inward spillovers are large but the outward spillovers are small.
Figure 2. Correlation of outward spillovers, size and export-driven growth
For Germany to play a stronger locomotive role would require sustained rebalancing of growth toward domestic demand. Stepped-up efforts to raise potential growth would contribute to that goal. In addition to creating incentives for greater domestic investment, higher permanent incomes would also sustain higher consumption growth. The current strengthening of domestic demand alongside external demand contributions to growth is encouraging, but it is too early to allow identification of a structural shift.
Our focus here on short-run growth dynamics should not be read to negate the broader and medium-term benefits accruing from, for example, German foreign direct investment abroad. We also set aside the normative question of whether a country should aspire to generate large spillovers – which can be a boon in times of cyclical expansion and a curse in times of cyclical downturn. Such considerations could suggest the need for a more normative metric of spillovers, one that incorporates considerations such as the sustainability of the expansion. Both questions are outside the scope of this analysis, but part of an ongoing research agenda.
The views expressed in this article are those of the authors and do not necessarily represent those of the IMF, IMF policy, or the Executive Board of Directors.
This and companion columns on fiscal spillovers and structural contributors to the current account surplus form background work for the IMF’s 2011 Article IV Consultation with Germany, which was concluded on 6 July 2011. An in-depth analysis of spillovers from the five largest systemic economies is also available in a series of spillover reports.
Bayoumi, T and Andrew Swiston (2009), “Foreign Entanglements: Estimating the Source and Size of Spillovers across Industrial Countries”, IMF Staff Papers, 56(2):353-383.
Danninger, S (2008), “Growth Linkages within Europe”, Chapter II in Germany: Selected Issues, IMF Country Report 08/81:13-22.
IMF (2007), “Decoupling the Train? Spillovers and Cycles in the Global Economy”, Chapter 4 in IMF World Economic Outlook, October 2007:121-159.
Poirson, H and S Weber (2011), “European Growth Spillover Accounting from Crisis to Recovery”, IMF Working Paper (forthcoming).
Posen, A (2011), “The Euro Payoff”.
Swiston, A (2010), “Spillovers to Central America in Light of the Crisis: What a Difference a Year Makes”, IMF Working Paper 10/35
1 The full list of countries includes Austria, Belgium, Canada, Finland, France, Germany, Greece, Ireland, Italy, Japan, the Netherlands, Portugal, Spain, Sweden, Switzerland, the United Kingdom, and the United States.