Who Are Those Similar Natural Gas Customers?

We try to be thrifty with our energy use around my house, so it’s natural for us to wonder how we are doing, compared with those in similar circumstances. To evaluate if there is something else we could be doing to use less. In the month to March 13, for example, we used 131 therms of natural gas according to our National Grid bill, for heating, hot water and cooking. Is that a lot or a little?

It seems that either National Grid sort of wants us to know, or there is some regulation requiring them to tell us. More likely the latter, based on the quality of information provided. According to our bill, “similar customers’ average usage high/low range” was 37 therms to 297. So what does that tell me about our 131? Nothing.

If you assume by the “range” they mean the “range” as a statistician would define it, that is the first problem. The range is the highest and lowest amount, not the typical amount. The highest and lowest amount could be the two National Grid customers with the most bizarre circumstances, or with malfunctioning meters. It may not be relevant to me. A better alternative would to be to provide the median and range one standard deviation out, so one could say about 2/3 or similar customers were between X an Y. Or perhaps an inter-quartile distribution, expressed in laymen’s terms as “the average was X and half of all similar customers were between Y and Z.”

Regardless of the measure, however, the 37 to 297 range makes me wonder how “similar customers” are defined. Does National Grid know the gas appliances each customer has, and base “similar customers” on that? If so, I want to hear about the customer who heated their home, cooked and provided hot water for just 37 therms.

More likely the range is for all homeowners, but National Grid could do better than that based on their own records. If you use no gas in the summer and lots of gas in winter, then gas is used for heating and not other purposes. If you use about the same amount of gas all year, then gas is used for other appliances and not heat. If you use some gas in summer but much more in winter, like we do, then gas is used for both other appliances and heat. National Grid may not know which other appliances, or how many, but it at least can distinguish customers by heat vs. other appliances.

A far better comparison could by created by cross-tabulating the National Grid customer list the city’s Real Property Assessment File. This would allow “similar customers” to be distinguished based on building type (detached house, semi-detached house, attached house, apartment building) and number of housing units in the building vs. the number billed. An adjustment could even be made based on square footage per housing unit. Comparing its own records by address with RPAD, National Grid would be able to figure out if the landlord accounts for all the gas use in an apartment building, or of the landlord buys much of it (for heat and hot water, for example) while tenants buy some of it (for cooking, for example).

The RPAD file has inaccuracies in both its measure of square footage and number of units, so it isn’t perfect. And of course none of this takes into account the number of people in each housing unit. But I’d bet that if the measure of energy use by “similar customers” were confined to NYC one-family rowhouses of about the same square footage that we have, using natural gas for heat and other appliances, the range would be smaller than 37 to 297, with the variance within one standard deviation smaller still.

Now to be fair, the City of New York does not provide ANY information on “similar customer’s” water use. And National Grid does provide more detailed information on energy use – in Upstate New York if you get a “home energy report.”

“National Grid has been sending out these home energy reports to selected customer since May, says Patrick Stella, a spokesman for the utility. He says about 130,000 National Grid customers in upstate New York have been getting the reports as part of a pilot program here (the National Grid program started in Massachusetts).”

“To be enrolled in the program, Stella says customers have to fit a set of criteria that enables to the company to compare their energy usage against that of 100 similar customers (fuel type, size of house) within a 2-3 mile radius. Customers were automatically selected for the program — and they can opt out if they don't want the reports. As of right now, Stella says customers can't self-enroll in the program. He says they'll examine how things have gone at the end of this year and plot their next steps after that.”

But it seems to me that with the availability of the RPAD file, this should be much easier to do in New York City, for everyone, with the results plotted on the bill.

In any event, based on the data cited in the link it seems as if we used slightly more gas in 2010-2011 than the most “efficient” homes up in the Albany area, where it is colder and houses are more likely to be detached. Perhaps we need more weather stripping.