France has seen a marked decline in its export performance over the last decade (see Figure 1). This has provoked concerns about the economy’s capacity to adapt to intensified globalisation in trade of goods and services. The country is not an exception in this respect, as virtually all advanced economies – with the notable exception of Germany – have experienced declines in market shares since the turn of the millennium. Despite France’s favourable sectoral specialisation and geographic orientation of trade, difficulties in selling abroad have led to deeper losses in market shares than other leading economies between 2000 and 2007.
Figure 1. France’s exports and export market1
1The export market measures the worldwide demand addressed to a country and is defined as domestic exports that would be expected if its market shares by volume remained at their value for the reference year, here 2005.
Source: OECD, Economic Outlook No. 86 database
Traditional export equations including indicators of export-market growth and price competitiveness fail to account for France’s relative inability to service foreign markets in the pre-crisis period (Villetelle and Nivat 2006, Cochard 2008, Blot and Cochard 2008).
Various supply-side explanations have been proposed in the literature to shed light on this puzzle:
- Vigorous domestic demand.
- Lower mark-ups due to head-to-head competition with Germany.
- Low non-price competitiveness of French export goods.
- Offshoring of entire production processes (especially in the automobile sector).
- Difficulties of French manufacturing firms to reach critical size for exporting due to various administrative and other barriers. Even these factors fail to fully explain France’s poor export performance.
An additional explanation for underperforming French exports is a housing boom: strong profitability in the construction industry, fed by rising house prices, might have diverted a portion of capital and labour resources away from export activity in France (Égert and Kierzenkowski 2010). Indeed, like the pattern observed in a number of other countries, house prices trended sharply upwards relative to manufacturing-sector production prices in the 2000s (Figure 2). Policies supporting home ownership were strengthened during the boom phase of the cycle and probably contributed to the buoyancy of the real estate market, especially given that the price elasticity of supply was low.
Figure 2. House prices and producer prices in the manufacturing sector
From 2000 to 2007, the French construction sector faced a very tight labour market, and this situation was very likely reinforced by the introduction of the 35-hour week. France had in fact one of the most severe labour shortages in this sector among the major EU countries (Figure 3). In the wake of significant labour shortages, the construction industry saw persistent upward pressure on wages; the basic hourly wage for construction workers rose faster than in other sectors. Diverging wage trends prompted a sectoral reallocation of labour, even though the skills demanded are not necessarily the same. Between 2000 and 2007, the construction industry’s share of aggregate employment rose by nearly one percentage point, with the sector accounting for a quarter of the French economy’s new jobs over that period. Nevertheless, the productivity growth in the manufacturing sector helped free up some labour with no detrimental effect on output. Likewise, the construction industry might have been also able to draw labour from the primary and tertiary sectors and not solely from manufacturing.
Figure 3. Labour availability as a constraint on activity1
1Share of respondents pointing to the shortage of the labour force as the main factor limiting building activity.
Source: European Commission.
Labour would not have been able to shift significantly towards construction had capital not followed suit. Construction offered the economy’s highest net operating profit margins at 28%, compared to an average of just 10.5%. INSEE data on the creation of enterprises seem to confirm the shift. Between 2000 and 2007, the number of firms in the construction industry grew by around one quarter and more than doubled in real estate activities but remained stable in manufacturing.
Empirical studies on OECD country data confirm that higher real estate prices can trigger an inter-sectoral reallocation of labour (Bover and Jimeno 2007). Conefrey and FitzGerald (2009) also provide evidence that the housing boom in Ireland and Spain could have diverted resources from more productive uses; housing investment rose to as much as 14% of GDP in the former and 9% in the latter in 2005.Econometric estimations lend support to the housing-boom hypothesis
Our research confirms that standard export equations estimated for France – comprised of indicators of export-market growth and price and cost competitiveness – have poor statistical properties and do not fully capture the pattern of export developments. We add a relative price variable intended to capture the movement of labour and capital between the manufacturing sector and the sectors of construction and real estate activities. This takes the form of the ratio of house prices relative to producer prices in the manufacturing sector. We find this new variable to be a robust determinant of French exports. It considerably improves the overall fit of the model, in particular between 2004 and mid-2007.The housing boom may have hurt exports in other OECD countries …
Our findings prompt further investigation. Does the negative correlation between a housing boom and export performance apply more generally in industrialised countries? To answer this question, we selected OECD countries that were not able to fully meet the world demand for their exports. These include Australia, Canada, France, the UK, Italy, Norway, New Zealand and the US. Figure 4 shows a growing deterioration of their export performance in the period immediately preceding the recent crisis. It is noteworthy that this pattern went hand in hand with significant increases in house prices relative to prices of manufactured goods. Estimation results carried out for this panel confirm that previous estimates obtained for France hold for a wider set of OECD countries; an increase in house prices relative to manufacturing goods prices can be associated with a decrease in exports.
Figure 4. Exports, world demand for exports and relative prices
Note: REL is the ratio of prices of houses relative to those of producers in the manufacturing sector.
Source: OECD Economic Outlook No. 86.
France’s external competitiveness problem seems to be acute especially when compared to Germany. Our results show that:
- Price and cost competitiveness indicators seem to be important drivers of German export performance in contrast to results obtained for France.
- The relative price of real estate is found to be statistically significant with a negative sign, just as for France.
This result implies that a decrease in the relative price of real estate helped German exports, just as its increase in France hindered French exports.Conclusions
The bottom line is that there is good reason to believe that a healthier construction sector has come at the expense of manufacturing strength and competitiveness, and that reversing some of the incentives to invest in housing should have a favourable payoff in terms of trade performance in the medium term. Indeed, some of the measures supporting the housing sector have been eliminated or weakened more recently in France. Yet important distortions remain, which may fuel property price inflation, at least in areas of supply constraints like the Paris region and the Côte d’Azur.1References
Blot, C and M Cochard (2008), “L’énigme des exportations revisitée : que faut il retenir des données de panel?”, Revue de l’OFCE, No. 106, July.
Bover, O and J Jimeno (2007), “House Prices and Employment Reallocation: International Evidence”, Bank of Spain Working Papers, No. 0705 and CEPR Discussion Paper, No. 6543.
Cochard, M (2008), “Le commerce extérieur français à la dérive?”, Revue de l’OFCE, No. 106, July.
Conefrey, T. and J. FitzGerald (2009), “Blowing Bubbles – and Bursting Them: The case of Ireland and Spain”, The Economic and Social Research Institute, Dublin.
Égert, B and R. Kierzenkowski (2010), "Exports and Property Prices in France: Are They Connected?", OECD Economics Department Working Papers 759, OECD Publishing.
Égert, B and R Kierzenkowski (2013), "Exports and Property Prices in France: Are They Connected?", forthcoming in The World Economy.
Villetelle, J P. and D Nivat (2006), “Les mauvaises performances du commerce extérieur de la France sont-elles liées à un problème de demande ?”, Bulletin de la Banque de France, No. 146, February.
1 These distortions include i.) a generally favourable tax treatment: capital gains of primary residences and secondary homes under specific conditions are not subject to capital taxation, ii.) home ownership is favoured over rental housing, as imputed rents are not taxed, while actual rental revenues are, iii.) renovation and maintenance work are subject to reduced VAT, one of the objectives being to support the building industry.