Given that the ACCC would not release its own dataset for analysis, Informed Sources provided its petrol price dataset, including average daily (simple average across the sites monitored by Informed Sources), average weekly (simple average of the average daily prices for that week) and average monthly (simple average of the daily prices for that month) prices for Perth, Adelaide, Melbourne, Sydney and Brisbane, to selected academics and consultancy groups. The selected academics included Don Harding of La Trobe University and me. This is the same data that Informed Sources had provided to the ACCC. That dataset does not include the Fuel Quality Premia, Mogas95 prices or the net taxes used by the ACCC in its analysis. I was able to calculate a similar measure to the ACCC measure but I was not able to replicate the ACCC measure. I calculate a Relative Price measure where the Average of the eastern state capital average prices is subtracted from the Perth average price. In order to confirm the ACCC analysis I estimate the following equation:

Relative Price_t = α + β₁FW_t + ε_t (3)

where

The table below shows the result of this exercise.

Numbers in parenthesis are p-values. Standard errors are Newey-West corrected.

The results are consistent with the original ACCC analysis. Before the introduction of Fuelwatch, it appears that the petrol price in Perth was about 3cpl higher than in the eastern capitals. After the introduction of Fuelwatch, it appears that petrol prices in Perth fell by 0.85cpl relative to the eastern capitals.

The ACCC argue that: “Of potentially greater concern is the possibility that something else entirely has driven the improvement in the relative price margin.” The ACCC, however, did not investigate the most obvious other factor — the entry of Coles into the WA fuel market in March 2004. Using the Informed Sources dataset, I investigated that possibility.

I calculate the following equation:

Relative Price_t = α + β₁FW_t + β₂Coles_t + ε_t (4)

where

Consistent with the ACCC analysis, I use the time period August 1998 to June 2007 for the empirical analysis. Results are shown in Table 2.

Numbers in parenthesis are p-values. Standard errors are Newey-West corrected.

The results are now very different from the ACCC analysis. The dummy variable associated with FuelWatch is now not statistically significant. The Coles variable is highly statistically significant and indicates that greater competition in the form of Coles entering the market caused the relative price of fuel to fall by about 1.7cpl. In addition, the adjusted R² are now much higher than before. This result is consistent with the ACCC initial expectation regarding the WA petrol market. As Brian Cassidy told the Senate estimates committee on 5 June 2008: ‘At the time, I readily agree with you — I did not actually say it in evidence — my feelings were that it was probably the entry of Coles that had the major impact on prices in WA, but I have been subsequently proven wrong’ (Senate Estimates Hansard, 5 June 2008: E7). The irony, of course, is that Brian Cassidy was not ‘proven wrong’; the ACCC did not report any analysis comparing FuelWatch and the entry of Coles into the WA market in its 2007 report. The analysis here supports his initial views.