Posted on 09 April 2007

 

Source: SEAISI

The 14^th meeting of IISI Economic Studies Committee was held from 1 to 2 March 2007 in Brussels.   As in the previous meetings of the committee, the agenda of the meeting included review and updates on the forecasting methodology and short range outlook overview.  There were also special presentations by committee members on their respective countries/regions.  The following is a summary of the meeting results and working documents of the committee. Please note that the worldwide forecasts results will only be available for the public after the IISI's Board of Directors' Meeting in New Delhi, India on 25-26 March 2007.

Medium Term Forecast

The objective is to forecast the trend for demand of finished steel products to the year 2010 and 2015.  The theoretical approach for the forecast is by using the standard commodity model.  In the model, commodities are considered intermediate goods, so that their consumption is determined as a derived demand from final production.  In a steel demand model, the steel consumption is determined from the output of the steel using sectors.    It is expected that the apparent steel use (ASU) and the output of steel using sectors are cointegrated and the relationship can be modeled by using error correction model (ECM).  

The proxy for the output of the steel using sectors can be the index of industrial production (IP) of the country.  However, this may be inappropriate for cases where the industrial production index does not account for the output in the construction sector.  A possible solution is the use of  Steel Weighted Industrial Production Index (SWIP).  The proposed sectors are construction, domestic appliances, electrical equipment, mechanical machinery, metal products, automotive, and other transport.  Another alternative solution is the adoption of Gross Fixed Capital Formation (GFCF) as a proxy for demand, and therefore production, for the steel using sectors.  In short, although the choice of the best proxy for the total production of the steel consuming sectors may depend on the characteristics of the economy of a country, data quality and availability is also important.  When several indicators are available, their forecasting performance can be compared. The following table shows some examples of SWIP index from several countries.

SWIP Index Comparison

Steel Consuming Sectors	Brazil	Japan	Venezuela	Egypt
Construction	30.0%	38.7%	40%	62%
Domestic Appliances	-	7.4%	-	8%
Electrical Equipment	4.2%	-	-	6%
Mechanical Machinery	34.5%	11.3%	22%	4%
Metal Products	5.9%	6.7%	34%	17%
Automotive	23.9%	27.9%	4%	3%
Other Transport	1.4%	8.0%	-	-

 
Short Range Outlook

The first approach is utilising SWIP methodology to estimate the real steel use of a country.  Since the trend in specific steel use can be estimated, the future real steel use can be forecasted.  The assumption is that there is a relationship between the activity in the individual steel consuming sectors and the trend in real consumption in a country.

Another approach is basically making use of expert opinion, based on the judgment of some experts in steel industry.  Typically the steel experts anticipate the short term future consumption (1-2 years) as they should have some knowledge on the steel consuming industries' dynamics.   The experts must have also considered the trends in macro economic indicators, especially the industrial production growth. These factors are weighted and evaluated by the experts.  For a start, the following table is the apparent steel consumption forecast for the region based on available information gathered via interviews and publications.

Apparent Steel Consumption                          ('000 tons)

COUNTRY	2004	2005	2006e	2007f	2008f
Indonesia	5,718	7,235	6,580	6,843	7,185
		26.5%	-9.1%	4.0%	5.0%
Malaysia	7,138	6,726	6,915	7,105	7,389
		-5.8%	2.8%	2.7%	4.0%
Philippines	3,041	2,679	2,813	2,954	3,101
		-11.9%	5.0%	5.0%	5.0%
Thailand	13,097	14,143	12,870	13,514	14,460
		8.0%	-9.0%	5.0%	7.0%
Vietnam	5,312	5,529	6,137	6,874	7,698
		4.1%	11.0%	12.0%	12.0%
TOTAL	34,306	36,312	35,316	37,289	39,834
		5.5%	-2.8%	5.3%	6.8%

 
Challenges in Forecasting South East Asian Steel Demand

The first challenge is the difficulties arising from the availability and quality of the required input data.  The forecasting models developed by IISI would normally require quarterly time series data and until today SEAISI does not compile data in a quarterly basis.  There is also a need to provide substantial details of national economic data especially on the output of steel consuming sectors. 

Another challenge is the differences in economic and industry structure and developments of the countries in the region.  Moreover, major changes in economic policies –which are common in emerging economies- would also add some complexity to any forecasting study. 

Since there is general consensus that reliable steel demand forecasts are important to both steel producing companies and governments -as they provide the basic assumption of any capacity expansion and industry related policies- we look forward to seeing a more active participation from SEAISI member countries in the near future.

 

«  Back