Does Price Responsiveness Exist in Cold Climates?

By Ahmad Faruqui[1]

There is a widely held view that customers in winter-peaking climates will not respond to any kind of time-varying rate, whether its static or dynamic.[2] The conventional wisdom is that customers value comfort more in the winter than in the summer and, thus, are more unlikely to adjust their thermostats downwards in the winter months than to adjust them upwards in the summer months (even though one can bundle up in the winter more easily than one can dress down in the summer).

Hydro Quebec, which is a winter peaking utility serving customers in a province with one of the coldest climates in North America, has shown that their customers do respond to time-varying rates. Frederic Pelletier and I co-authored an article in The Electricity Journal on that finding.

Now, results have come in from a two-year pilot with time-varying rates that was carried out by Nova Scotia Power[3], one of the maritime provinces in Canada. The impact evaluation was carried out by Econoler.

First, here’s a quick snapshot of monthly temperatures in Halifax, Nova Scotia, just to make the point that winters are really cold there.

The pre-pilot rate was the standard residential rate, with a fixed monthly charge of $19.17 a month (~USD $14.5) and a flat energy charge of 15.744 cents per kWh (~USD 11.8 cents per kWh). Both time-of-use (TOU) and critical-peak pricing (CPP) rates were tested in the pilot over two years.

The TOU rate only applied in the winter months. It featured peak prices during the morning and evening hours, which is typical of winter peaking utilities. In all other months, a 30% lower flat rate was applied.

The TOU rate had a 2:1 ratio between peak and off-peak rates. It yielded a reduction of 9-10% in winter peak demand. The critical-peak pricing rate had a 10:1 ratio between critical-peak and off-peak rates. It yielded a reduction of 28% in winter peak demand on critical days.

How do these results compare with results elsewhere?

The response to TOU rates is twice as high as I would have expected from the meta-analysis of nearly 400 tests of time-varying rates around the globe that are embodied in Arcturus, a database that I have been working on from the beginning of my rate design career in 1979. Most of the data in Arcturus come from summer-peaking utilities. The Nova Scotia results are a big surprise, especially given the skepticism about the viability of time-varying rates in cold climates.

The response to CPP rates is ten percentage points higher than the 18% that Arcturus would have predicted.

For reference, here are the results from Hydro Quebec’s implementation of CPP and PTR rates.

Nova Scotia Power’s customers on TOU rates achieved a modest annual savings of 4% in their electric bills. Interestingly, customers with electric heating saved 2.2% while non-electric heating customers saved 9.5%. That’s the opposite of what I would have expected. In most summer peaking climates, homes with central air conditioning respond a lot more than homes without central a/c. For example, in California, homes in cooler zones, such as San Francisco, the coastal areas and the mountainous areas, respond a lot less than homes in much warmer zones, such as the Central Valley. That was found to be the case in the Statewide Pricing Pilot in the early 2000’s and has been reinforced subsequently by other pilots.

In Nova Scotia, customers on CPP rates saved 14% on their annual electric bills, a robust amount. Usually, when customers enroll in a full-scale rollout of time-varying rates, they are looking to save something on the order of 10%. Otherwise, it’s not worth the bother.

That’s why most opt-in offerings of TOU rates do not get enrollment rates much higher than 15%, as discussed further in the chapter on time-varying rates that I have written for a forthcoming Handbook of Regulation. [4]

In Nova Scotia’s pilot, higher income customers reduced their peak usage by less than lower income customers, another surprising result. Usually, higher income customers have more discretionary load in their homes.  

How did the customers modify their usage patterns? Most customers reported rescheduling their laundry, dishwashing and cooking activities in response to the time-varying rates. Some 45% reduced or shifted their space heating load outside of the peak hours and 29% pre-heated their homes before peak hours.

APPENDIX: AN INTRODUCTION TO TIME-VARYING RATES

Time-varying rates come in many forms. The main ones are described below and illustrated in the tables and graphs that follow.