While the industry can debate what Australia’s final export volumes might look like into the end of this decade, the price of LNG in Asia over coming years will be critical. Given that long-term LNG contracts in Asia tend to be based on formulas linked to the lagged price of oil, I use historical data to create a forecast model based on Brent prices.
Creating such a model can be difficult. Some contracts can be complex in nature. It has been common for instance for contracts to have formulas that have caps and floors. The impact of these ‘S-curves’ would tend to be more obvious at higher and lower prices to varying degrees. The nature of contracts will also change over time as demand for LNG in Asia changes. All this can make it difficult to isolate and create a ‘one-size-fits all’ model.
Chart 1 plots Westpac’s preferred models using the following methodology:
- P =C+S*Brent (t-x)
Where P is the price of LNG in Japan in $/MMBtu
C is a constant expressed in $/MMBtu
S is the slope or coefficient
x is the average price of Brent lagged by 3 to 6 months.
Westpac uses a constrained regression model to plot a best-fit line based on the assumption that the ‘S-curve’ kicks in at fixed points, e.g. AU$60 and AU$90. The model run two models, one that has gradients fixed above AU$90 and below AU$60 and one that has non-fixed slopes.
Armed with the above equations, we can attempt to forecast prices based on historical and forward Brent prices.
Chart 3 suggests the model appears to do a fairly good job of forecasting Australian LNG prices in Japan over the last 10 years, and it has been correctly forecasting the 50 per cent drop in LNG prices this year.
While it has arguably over-forecast the recent drop, the current shape and level of the Brent curve backs up the earlier point that the industry is close to passing the worst for Australia’s LNG export prices.
The export boost to Australia