Sunday, May 10, 2009

EMU Economic Indicators Preview (Week of 11 to 17 May 2009)

  • German GDP (Q1 2009): down
  • EMU GDP (Q1 2009): down
  • EMU industrial production (March): down

German GDP will probably have plummeted quarter-on-quarter in Q1 2009, just as output in the producing sector did. The previous quarters' data might be revised slightly. On 15 May, Destatis (the German Federal Statistical Office) will publish its “flash release” on Q1 German GDP; a detailed breakdown of the components will follow on 26 May. However, no demand component is expected to have contributed positively to overall GDP growth, except for government consumption. French, Italian and EMU GDP are all likely to have contracted significantly in Q1 too.

French, Italian and EMU industrial production will probably have continued declining in March, because most of the correlated indicators deteriorated.

Final HICP inflation in the eurozone will probably be confirmed at 0.6% in April; this would correspond with a monthly inflation rate of 0.4%. As from June, inflation rates are expected to dip into negative territory and remain there for several months.

Methods: Forecasting German GDP

Our forecasts for German GDP are mainly based on regression analysis (“econometrics”). Here, seasonally and working-day adjusted growth rates qoq (GDPG) are explained by:

  • German output in total industry excl. construction, seasonally and working-day adjusted, in % qoq (IP),
  • German output in the construction sector, seasonally and working-day adjusted, in % qoq (CP),
  • German retail trade excl. trade and repair of motor vehicles and motorcycles, seasonally and workingday adjusted, in % qoq (RT), and by
  • German changes in inventories, seasonally and working-day adjusted, growth contribution to GDPG in percentage points qoq (CI), lagged by 1 quarter.

The coefficient of determination R2 indicates that this model explains more than 74% of the total variation in German GDP growth, which is a very high proportion when estimating growth rates quarter-on-quarter. The explanatory variables are highly significant, as their absolute t-values exceed 2. The t-values are Newey-West heteroscedasticity and autocorrelation consistent. The Breusch-Godfrey test indicates no autocorrelation up to the 4th order (n*R2=1.566, probability=0.815).

As the chart shows, this model has performed very well during the last three years, as opposed to before that. It even signals later revisions quite reliably.


This report has been prepared by BHF-BANK Aktiengesellschaft on behalf of itself and its affiliated companies (together "BHFBANK Group") solely for the information of its clients.

The information and opinions in this document are based on sources believed to be reliable and acting in good faith, but no representation or warranty, express or implied, is made by any member of the BHF-BANK Group as to their accuracy, completeness or correctness. Opinions and recommendations are given in good faith but without legal responsibility and are subject to change without notice. The information does not constitute advice or personal recommendation, for which the duty of suitability would be owed, but may facilitate your own investment decision. Moreover, you should seek your own advice as to the suitability of an investment matter mentioned herein. Investors are reminded that the price of securities and the income from them can go down as well as up and that the past performance of an investment or a market is not necessarily indicative for future results.

This document is for information purposes only. Descriptions of any company or companies or their securities mentioned herein are not intended to be complete, and this document is not, and should not be construed as, an offer to sell or solicitation of any offer to buy the securities mentioned in it. BHF-BANK Group and its officers and employees may have a long or short position or engage in transactions in any of the securities mentioned in this document, or in any related securities.