How should time-varying rates be structured for households?

For more than four decades, going back to 1978 when MIT’s Fred Schweppe published a paper[1] on homeostatic control, a few academic economists and engineers have pushed hard for implementing hourly, real-time (RTP) pricing. With time, the concept has become even more arcane, with some pushing for the application of the RTP down to the distribution level. They call it locational RTP.

Sometime in the 1990’s, I had the pleasure of meeting Fred Schweppe at an IEEE conference in Washington, DC. The hall was full and he stood out. There is no doubt that Schweppe was a force of nature.

In the late 1980’s, I was speaking on time-of-use (TOU) pricing at a meeting of the Midwest Economic Association, maybe in Indianapolis. Professor Westfield of Vanderbilt University and I were on a panel. When it came my turn to present, I noticed there was a man in the front row who was sitting with eyes closed. Was he asleep? When I had finished speaking, he put a very interesting question to me. Wow, he had been listening all along. Later, when I went down to chat with him, I discovered he was the legendary William Vickrey, who had written a landmark paper on dynamic pricing seven years before Schweppe’s.[2] In 1996 he won the Nobel Prize in Economics.

Fast forward to today. A few practitioners of the art of pricing, notably in California, have joined forces with the academics, pushing hard for RTP. I used to be in that camp. In 2006, I had the same dream. I published an article that laid out my “conceit,” as Robert Earle, one of the sharpest economists I had the pleasure of working with at Charles River Associates put it.[3] In 2019, I revisited the theme. The occasion was EEI’s Rate and Research Committee meeting in San Diego.[4]

Yet, I could not help but notice that the concept garnered scant attention among households. At some point I stopped talking about RTP. I had finally realized, to my chagrin, that just about no household wanted it. So why was I wasting time advocating for an idea that was born in the classroom and that had no traction outside of the classroom?

I noticed that only a single state, Illinois, was offering RTP to residential customers. More than 15 years later, only 2% of customers had selected it, despite claims that an overwhelming majority of customers could save substantial sums of money by switching to RTP.

It’s clear that RTP, idealized in the halls of academe, has failed to gained traction with households. Can time-of-use pricing, possibly augmented with critical-peak pricing (CPP) or peak-time rebates (PTR), provide a good second-best solution?

Even TOU pricing won’t be a slam dunk. Despite decades of being offered, it remains a hard sell. In 2021, according to the US Energy Administration, only 9% of residential customers were on simple TOU rates, even though 69% of households had smart meters. That was double the percentage of 2018, which was not much changed from the percentage in 2013. Was the glass half empty or half full?

Are we on the cusp of a TOU revolution? It's too early to say.

SMUD was probably the first utility to make TOU the default rate. The transition was done very carefully. It was preceded by a very well designed pilot that included a track for opt-in participation and a track for opt-out participation. Then came a three-year customer education and outreach campaign. The rate features a significant discount during the off-peak period. Savings opportunities are substantial. I cannot say enough good things about SMUD’s deployment.

The three investor-owned utilities are in the process of rolling out a mild TOU rate as the default in California. No one will save much on that rate and a lot of people will get annoyed. I never understood the logic. Struck me as a check-list type of formality.

Colorado's Xcel Energy is deploying a much better TOU rate as the default as it deploys smart meters. A friend informed me that a cooperative is also about to do the same. Of course, Fort Collins has implemented mandatory TOU rates. It’s small in size and it’s a municipal utility.

The two investor-owned utilities in Michigan have already done so. In fact, the only other rates that you could opt-out to are also TOU rates. So, Michigan becomes the first state to make TOU mandatory!

And, in Missouri, the two investor-owned utilities are about to deploy the most aggressively designed TOU rate (with peak prices that are four times higher than off-peak prices; that ratio of 4:1 is higher than seen anywhere else) as the default tariff in October.[5] I hope they succeed. All eyes are on them. We are looking for good news, not a big bang.[6]

Where is RTP? Nowhere to be seen? Well, it's there in Illinois. Two percent (2%) of customers have taken it even though studies have shown that it can benefit as many as 90% of customers.

What explains the reluctance to take RTP? Apathy? Repugnance? Anger? Ethics? Those are the words you hear when you mention RTP to customers. In fact, even a simpler form of dynamic pricing, CPP, evokes similar sentiments. In 2010, I was asked to speak at a conference at Rutgers University. It was on the ethics of dynamic pricing. Who sponsored it? Not the economics department. Not the electrical engineering department. It was the department of moral philosophy. That’s how touchy the subject is, just like a live wire. 

With that background, it’s worth reading a new paper[7] that compares the merits of TOU+CPP with RTP. Surprisingly, the authors are academics, including the renowned Paul Joskow of MIT. As best as I can tell, It's the first paper from academics that makes the point I have been making for a few years now: The perfect is the enemy of the good. 

 

Here's the paper’s abstract.

Electricity Retail Rate Design in a Decarbonizing Economy: An Analysis of Time-of-use and Critical Peak Pricing 

Tim Schittekatte, Dharik Mallapragada, Paul L. Joskow and Richard Schmalensee

Abstract [emphasis added by me]

Currently, the main component of most U.S. consumers’ electricity bills is based on a constant price per kWh consumed. As intermittent renewable resources and flexible loads that can be shifted within days (such as electric vehicle charging) gain prominence in the electricity system, the efficiency gains to be realized from basing bills instead on wholesale spot prices increase. 

There is little political support for this change, however. 

We focus on second-best alternatives: time-of-use (TOU) rates and critical peak pricing (CPP). We introduce alternative assessment criteria that focus on intra-day load shifting. Using historical data, we find that TOU rates can reasonably replicate the intra-day load-shifting incentives provided under spot pricing. Thus, TOU rates, especially when complemented with CPP involving load control during infrequent scarcity price events, can be considerably more socially valuable than previously estimated.