Short Term Update 04-14 : Special Topic - Domestic demand patterns in Belgium since the mid-nineties: more French than German?
In a previous Special Topic (see STU 4-11), we examined the performance of Belgian GDP relative to its three main trading partners since the onset of the financial crisis. This time we put recent domestic demand patterns into perspective by comparing, over a longer period, the evolution of private consumption and business investment in Belgium to those in France and Germany. In addition to the purely descriptive analysis, we also try to identify, through the estimation of co-integrating equations, the underlying factors behind the developments of both demand components under scrutiny. We investigate in particular two sub-periods: the decade preceding the financial crisis (1998-2007) and the following five years (2008-2012).
Evolution of private consumption and business investment1
Between 1995 and 1999, private consumption in Belgium and France grew on average at a very similar pace. It became more dynamic in France in the following years, reaching a cumulative growth differential of about 9% in 2007. This gap has been progressively reduced to around 5% since the financial crisis. In Germany, private consumption grew much more slowly until the outbreak of the crisis; however, the maximum differential of 16% with Belgium has been reduced by 4 %-points over recent years.
Business investment had very similar evolution in France and Belgium until the crisis, but has stood up somewhat better in Belgium in recent times. Following the reunification boom, German business investment growth has been much more subdued over the whole period, with no recent signs of convergence despite the recovery.
To be complete, note that the story for exports (not dealt with this Special Topic) is exactly the opposite, with France and Belgium both recording very low performances compared to Germany.
Key factors behind these evolutions
In a recent working paper, we investigated econometrically the underlying factors behind private consumption and business investment developments2. More specifically, we estimated for the three countries under review a co-integrating equation postulating an equilibrium relationship relating both demand components to a selection of medium-term determinants.
According to our estimates, there is evidence of a strong link between consumption and real disposable income for the three countries. The income elasticity is slightly above 1 in Germany and stands at about 0.85 in France and Belgium. We found no indication that financial wealth has a statistically significant influence on consumption in Germany and it only has a small effect in France and Belgium. Real house prices play a (limited) role only in Germany and are negatively signed, which may be explained by the low levels of house ownership in this country. Higher house prices may also imply a larger mortgage debt service. Finally, a decline of 1 %-point in the German unemployment rate is found to improve consumption by about 0.3% in the medium-term.
As postulated by the accelerator model, business investment behaviour is largely driven by output developments. The medium-term elasticity to private value added is very similar in all three countries and stands at 1.4. While non-significant for Germany, both France and Belgium post elasticities to the cost of capital that are slightly below unity. The estimates suggest an almost one-to-one relationship between investment and firms’ profit margins in France, but the effect on the former is clearly more limited in Germany (0.6) and even more so in Belgium (0.2). The ratio of total liabilities to equity is negatively signed in the three countries, suggesting that elevated corporate leverage is associated with weak investment. A 1% increase in this ratio dampens investment by 0.2% in France and Germany and up to 0.3% in Belgium. In addition to common determinants, other country-specific factors have acted as drags on investment in both Belgium and Germany, which are captured by a negative time trend.
Contributions of explanatory variables
Multiplying the average annual growth rate of each explanatory variable by its regression coefficient allows its contribution to private consumption or business investment growth to be computed. The contributions presented below are divided into two sub-periods: the decade preceding the financial crisis (1998-2007) and the following five years (2008-2012).
For the period 1998-2007, differences in private consumption growth are mainly related to real disposable income, which increased significantly more in France and Belgium than in Germany. Financial wealth effects added somewhat to the positive growth differential of France and Belgium vis-à-vis Germany. During and after the crisis, private consumption growth amounted to around 1% per year on average in Belgium and Germany. They clearly outperformed France, which suffered from less dynamism in disposable income. Despite the higher elasticity in Germany, disposable income posted a higher contribution in Belgium. However, German consumption received an additional boost from the decline in its unemployment rate, while it was pulled down slightly by a negative income effect stemming from the rise in house prices.
About half of the lower growth in business investment recorded in Belgium (3.7%) compared to France (4.8%) during the period 1998-2007 can be explained by less robust growth in private value added (recall that elasticities in both countries are almost identical). The large negative investment growth differential experienced by Germany vis-à-vis both neighbours (respectively 1.1 and 2.2 %-points) cannot be explained by the variables included in our equations. One important factor put forward by several authors to explain the sluggish capital formation in Germany is that corporations started to redirect investment from the domestic economy to international locations, off-shoring production and creating supply chains, particularly in low-cost Eastern Europe upon the eastward expansion of the European Union.
While the negative investment growth in France (-0.8%) during the period 2008-2012 is entirely explained by the stagnation in value added and the deterioration in financing conditions and profitability, a similar performance for Belgium is only poorly described by the traditional determinants. Industry offshoring, for instance in the automotive industry, is likely to have weighed on Belgian investment during this period. We also found some empirical evidence that both confidence and uncertainty effects have contributed to the drop in the investment rate over 2009-2010 in Belgium (for details, see the working paper). According to the explanatory variables identified, German business investment should have stagnated since the outbreak of the crisis, but in fact it declined even more (-1.1% on average) than in France and Belgium.
To sum up, during the decade preceding the crisis, domestic demand patterns in Belgium were somewhat less dynamic than those in France but clearly more vigorous than in Germany. Since the crisis, similar weak performances in business investment have been recorded in all three countries, while Belgian and German private consumption have shown extra resilience due to more favourable developments in disposable income or on the labour market.