As part of his recent two-part investigation into NW Natural, journalist Chris May relied on company data and publicly available information to calculate two figures. In the interest of full transparency, here is an explanation of how he crunched the numbers.

“95 percent of all excessively inaccurate meters ran ‘fast’”

Gas meters that NW Natural suspects are inaccurate but have yet to be tested are called “non-conforming meters.” When trying to understand the proportion of meters that run fast, Chris started by looking for non-conforming meters.

The company’s meter test data from 2019 reported that 10,503 non-conforming meters were tested, and 2,296 were reported as inaccurate. Of those, 2,193 tested “fast” while 103 tested “slow.” The “fast” meters make up 95 percent of the total inaccurate meters NW Natural reported in its meter test data.

Chris also looked at a 2021 communication from NW Natural to OPUC, which notes that “from 2019 to date [July 27, 2021], NW Natural has processed 9,218 bill credits,” which are issued to customers whose meters run fast, compared to “140 commercial and 222 residential” re-bills for slow meters. He added the number of fast meters to the number of slow meters for a total of 9,580 inaccurate meters. The “fast” meters make up 96 percent of the total inaccurate meters NW Natural reported to OPUC in July 2021. Because company bill credits “do not break down whether the meter was a non-conforming meter or any other type of meter,” according to NW Natural, this figure includes non-conforming meters and meters tested for other reasons.

“The utility could have been overestimating its gas costs by up to $4.3 million from 2002-2004”

Chris looked at NW Natural’s filings with the Securities and Exchange Commission (SEC), specifically its Form 10-K from 2004 and 2005. 

He looked at the annual total gas sales (in therms) and the cost per therm reported by NW Natural to the SEC for 2002, 2003, and 2004 in the company’s 2005 Form 10-K, as well as the percentage of unaccounted-for-gas reported in NW Natural’s 2004 Form 10-K filed with the SEC, which includes figures for 2002 and 2003 as well.

Annual total gas sales (in therms) reported by NWN in its 2005 Form10-K:

2004: 742,352,000

2003: 685,198,000

2002: 680,085,000

UAG % reported by NWN in its 2004 Form 10-K: 

2004: 0.51%

2003: 0.55%

2002: 0.75%

Cost per therm reported in by NWN in its 2005 Form 10-K:

2004: $0.57

2003: $0.47

2002: $0.51

He then calculated what the difference in gas costs would be had NW Natural reported unaccounted-for gas losses of 1 percent, as alleged by Dye, compared to gas costs based on the unaccounted-for-gas percentages reported in NW Natural’s 2004 10-K.

(Total gas sales * 1% UAG) – (Total gas sales * 2004 10-K %) = UAG Difference in therms

2004: (742,352,000 * .01) – (742,352,000 * .0051) = 7,423,520 – 3,785,995 = 3,637,525

2003: (685,198,000 * .01)  – (685,198,000 * .0055)  =  6,851,980 – 3768589 = 3,083,391

2002: (680,085,000 * .01)  – (680,085,000 * .0075)  = 6,800,850 – 5100638 = 1,700,212

Finally, he took that “UAG Difference in therms” figure for each year and multiplied it by the cost per therm NW Natural reported in its SEC filings to determine the amount by which the company would be overestimating its gas costs.

Difference in cost = (UAG difference in therms * annual cost per therm in SEC filings)

2004: 3,637,525 *  $0.57 = $2,073,389

2003: 3,083,391 * $0.47 = $1,449,193

2002: 1,700,212 *  $0.51 = $867,108

Total: $4,389,690