Our demand analysis is complemented by ‘bottom-up’ sectoral analysis, which focus
on changes in end-user demand – for example, the impact of the switch from
gasoline to diesel vehicles in Europe, the effects of the rapid expansion of
petrochemical capacity in Asia and the Middle East, or the growing use of natural gas
and coal for power generation in Europe. Although this approach provides a valuable
comparison with the main ‘top-down’ assessment, its results must be interpreted with
caution. For example, will there be sufficient feedstock supply (predominantly
naphtha) to sustain China’s petrochemical expansion plans? Will this expansion
result in a regional petrochemical surplus, leading to the closure of less efficient
plants? Global demand is predominantly driven by two primary forces, price and income. If
the assumptions underlying those variables are wrong, then the output of the model
will be affected. Similarly, the robustness of the historical data underpinning the
model is important. Finally, substitution issues can work in many ways, leading to
additional modelling complications.
Prices
When constructing our medium-term forecasting model, a price assumption was one
of the harder concepts to incorporate. The dynamics of the oil market have changed
considerably over the past five years and we are unaware of any formal price model
that predicted a rise in the oil price from $25 to $90/bbl over the past four years. The
seemingly robust relationship between oil stocks in the OECD and prices appeared
to break down, Additional inputs such as refining capacity, spare upstream capacity,
investment flows, natural gas prices and the term structure changes between spot
and forward prices have all had an impact.
