|
|
|
|
|
|
November 19, 2001
Summary
Malaysia national-monthly-average rainfall in the third quarter of 2001 dipped below normal for only the second time in the last 10 quarters. Additionally, excess precipitation in November 2000-January 2001 likely reduced palm pollination prospects. As high yields are typically associated with above-average precipitation, the reduced third quarter has reduced palm oil yield prospects.
Unusual Precipitation Patterns
Rainfall in the third quarter of 2001 dipped to 161 millimeters, just below normal for only the second time in the last ten quarters. In contrast, excessive precipitation of over 300 mm per month which occurred in November and December of 2000, and January of 2001, causing pollination problems and having a negative effect on output in the third quarter of 2001. Harvest of fruit bunches takes place approximately 6 months after pollination, so the effect of the excessive rainfall around the New Year has probably ended.
Model Indicates Reduced Yields
As a result of reduced rainfall levels in Malaysia, yield for 2001/02 (Oct. - Sept.) is forecast by our Malaysia Rainfall Regression Model (MRRM) to be just 3.79 tons per hectare. This is slightly above average, but below the 4.06 tons per hectare seen in 2000/01. Production reported by the Malaysian Palm Oil Board for the third quarter came in at 2.98 million tons of crude palm oil, down from 3.47 million forecast by the MRRM. Despite this lower-than-forecast output, yield for the entire 2000/01 year at 4.06 tons per hectare was the highest on record going back to 1982/83. The high yield is attributed to consistently-above-average rainfall since the third quarter of 1998.
Using estimated area of 3.31 million hectares, the MRRM projects output at 12.5 million tons of crude palm oil for the 2001/02 year. The MRRM projection is above the official USDA forecast for 2001/02 which is set this month at 12.2 million tons. Information from other sources indicates that output in the next several months may be below the level projected by the MRRM.
Note: The linear regression model uses rainfall lagged 3 quarters, 1 year cumulative rainfall lagged 6 quarters, and time as independent variables regressed against yield. Because of the characteristics of linear regression, the model continues to increase projected yields even as rainfall becomes excessive; thus, the MRRM tends to overstate yield when national-monthly-average rainfall lagged 3 quarters is above 300 mm, and understates yield when rainfall is optimal. Official USDA estimates for Malaysian palm oil forecasts, as well as for other commodities, are determined by using convergence of evidence from a wide variety of information sources.
