8/16/09

Prediction of price of gas at the pump: August-December 2009

Four months ago we made a prediction of price of gas at the pump [1]. In essence, we drew a straight line shown in Figure 1 as the future gas price for the period between March and December 2009. The line represents the difference between core CPI and the motor fuel index. The line says that the price for gas will be growing faster than the price of all goods and services less food and energy (the definition of core CPI). A midterm review is a natural step in tracking the following prediction [1]:

In March 2009, the difference was at the level of +45, i.e. much higher than the level predicted by the new trend. As happened in the past with numerous individual price indices [9,10], such a strong deviation (one might call it “dynamic overshoot”) should be compensated in the near future. Without loss of generality, we have restricted the recovery to the trend by the end of 2009. As a result the index for motor fuel should growth by 90 units during the next 9 months, or by 10 units per month. Red filled circles represent the evolution of the difference from April to December 2009. In 2010, the difference may undergo an overshoot in the opposite direction with additional rise in the index for motor fuel.
Translating indices into prices, the rise in the difference by 90 units (from 173 in March to 263 in December) means an increase in price by 50%. Therefore, it is very likely that the price for motor fuel in the beginning of 2010 will be 60% to 70% larger than in March 2009 due to the overshoot.

Having the latest estimates of the core CPI and the motor fuel index, as published by the BLS, we have calculated the difference for July at +13 units of index instead of predicted +5. Figure 2 displays relevant curves and confirms that the overall trend in the difference holds. Considering high volatility in relevant price indices one could not expect a one-to-one correspondence between the observed and predicted curve, and a small delay observed in June will be compensated in August or September.
We have been also reporting the evolution of crude oil price, which obviously affects the price of motor fuel. Despite high resemblance, these prices have no one-to-one correspondence and it is instructive to model them separately. So, we will continue tracking gas price at the pump. The delay in June allows to predict a surge in the price in August and likely in September. We also retain the mid-term prediction, i.e. the price in December 2009 and in the beginning of 2010, at 70% higher than in March 2009. Bearing in mind our prediction of oil price at $100 by December 2009, we might put the motor fuel price higher with probability increasing in time, since our predictions were accurate.

Figure 1. The difference between the core CPI and the index for motor fuel. Red filled circles predict the evolution of the difference between March and December 2009. Total increase in the difference is +90 units of index or +50%: from 173 in March to 263 in December. Solid red line represents the “mirror” trend for that between 2002 and 2008, which is shown by solid black line.

Figure 2. The difference between the core CPI and the index for motor fuel. Red filled circles predict the evolution of the difference between August and December 2009.


References

[1] Kitov, I., Kitov, O., (2009). A fair price for motor fuel in the United States, MPRA Paper 15039, University Library of Munich, Germany, http://mpra.ub.uni-muenchen.de/15039/01/MPRA_paper_15039.pdf

No comments:

Post a Comment