This post is an ode to consumer demand and its importance to determining value in electricity.
In Book V, Chapter II, paragraph 7 of his seminal Principles of Economics in 1890, economist Alfred Marshall described the market interaction between supply and demand as two blades of a pair of scissors.
We might as reasonably dispute whether it is the upper or the under blade of a pair of scissors that cuts a piece of paper, as whether value is governed by utility or cost of production. It is true that when one blade is held still, and the cutting is effected by moving the other, we may say with careless brevity that the cutting is done by the second; but the statement is not strictly accurate, and is to be excused only so long as it claims to be merely a popular and not a strictly scientific account of what happens.
Market value is a result of the interaction of supply and demand, and it's an emergent result, not one that is known deterministically in advance. This interaction allows price discovery, the process of learning what that value is, to occur. That value is not known in advance, to anyone.
We often analogize from consumer product markets when thinking about supply/demand interactions, but the electricity industry is very different from consumer product industries. From its origins in the 1880s it's always been a very supply-driven industry, not surprising when you think about the large fixed costs, economies of scale, network infrastructure nature of the industry.
The electricity industry is so supply focused because it is a critical infrastructure that must be reliable and available all the time. Without enough supply to meet demand, blackouts or brownouts can occur. Electricity cannot be stored in large quantities, so it must be generated as it is needed. This means that there must be enough generating capacity to meet peak demand, which can be significantly higher than average demand. The electricity grid is a complex system that is difficult to manage. A number of factors can affect the grid, such as weather, equipment failures, and cyberattacks. These factors can cause outages, so it is important to have enough generating capacity to cover unexpected events.
In case it's not obvious, that last paragraph is the product of Google Bard, not me. This is the received wisdom about the supply focus in electricity, and it's correct as far as it goes. But notice the implicit assumption that pervades the language: "enough supply to meet demand", "enough generating capacity to meet peak demand": supply adjusts to meet fixed demand. Demand has to be met, I flip the switch and the light goes on. So much of how we think about the transactions, the organizational structure, the business model, the regulatory institutions in electricity, all of the economics involved in understanding this industry, is predicated on this assumption.
This assumption limits our ability to imagine different approaches to electricity in a changing economic and technological context. It's an assumption that's been obsolete for a while, and has become increasingly obsolete with digitization and automation over the past 15 years. Digital consumer-facing technologies make it easy and convenient for consumers to change their quantity demanded, either through human action in an app or through automation, in response to some signal. In many instances we can time-shift our consumption (and should be compensated for doing so), and the thermal mass of buildings and water heaters even means that we can reduce quantity demanded without a loss of utility to the end user.
With these technologies, both supply and demand can be flexible. The marginal value/marginal cost of adjusting one or the other changes over time according to production costs, consumer utility, and system conditions. The only way to learn at any point in time which adjustment has the highest marginal value/lowest marginal cost is through a market process of price discovery.
Source: Marshall, Principles of Economics, Book III, Chapter IV, "The Elasticity of Wants"
Part of the economic justification for assuming that demand must be met is that its price elasticity of demand is low, that consumers are not responsive to changes in price. Typical estimates for residential consumers for short-run price elasticity are small, around -0.3, which means that a 10% increase in price leads to a 3% reduction in quantity demanded. Long-run elasticity, once people have time to observe the prices they are charged (and billed for well after the fact) and adjust some of their devices, is closer to -1 and therefore a larger response. Bernstein and Griffin (2006) estimate short-run residential elasticity at -0.2 and long-run at -0.7. Burke and Abayasekara (2018) estimate that across all types of customers (industrial, commercial, residential), short-run price elasticity of demand is -0.1 and long-run elasticity is -1. Feehan (2018) uses a natural experiment of an "abrupt, substantial, and permanent" change in electricity price to perform a difference-in-differences estimation of residential price elasticity; he finds a long-run elasticity of -1.2. His introduction gives an indication of how wide the range of estimates are:
Estimates of its price elasticity are plentiful and diverse, and reflect both differences in space and time but also in estimation techniques. In a frequently cited contribution, Espey and Espey (2004) carried out a meta-analysis of price and income elasticity estimates from 36 studies published over the period 1947 to 1997. From the 123 estimates that they analysed, short-run price elasticities ran from −2.01 to −0.004 with a mean of −0.35; and 125 estimates of long-run price elasticity fell in the range from −2.25 to −0.04 with a mean of −0.85. Differences in econometric techniques may explain some of the variation, but even with the same methodology, a wide range of estimates can be obtained. For example Krishnamurthy and Kristöm (2015), using a common methodology, obtained a range of price elasticities between −0.27 and −1.4 for a set of 11 OECD countries.
These elasticity estimates reflect the conventional analog mechanical household technologies that require attention and manual adjustment. They also reflect the extensive economic regulation of pricing and the pervasiveness of fixed rates in the typical residential utility tariff. These two dimensions, technological and regulatory, go hand in hand – without technologies to respond to dynamic prices in a timely fashion, consumers would face considerable bill uncertainty at the end of the month if they had dynamic pricing but had to rely on manual controls without current price information.
Digital home energy technologies change that relationship and change the possible value opportunities for consumers who choose to vary their consumption. Smart thermostats allow users to control and schedule their HVAC systems remotely. By connecting to the internet and leveraging sensors, weather forecasts, and user preferences, these devices cam optimize energy use based on occupancy patterns, weather conditions, and energy tariffs, and can automatically adjust temperatures to reduce energy consumption during peak periods or when the home is unoccupied. Home energy monitors provide real-time data on energy usage at a granular level. By visualizing consumption patterns and identifying energy-intensive appliances, consumers can make informed decisions about when and how to use electricity. These systems may offer recommendations for energy-saving practices, such as identifying appliances that are left on unnecessarily or suggesting more efficient alternatives. Even in a regulated tariff setting, demand response programs can change how responsive consumers are to price signals. Through digital platforms and smart devices, consumers receive signals or price incentives indicating peak demand periods. By adjusting their electricity usage accordingly, such as running appliances during off-peak hours or reducing non-essential energy consumption, consumers can help balance the electricity grid and potentially receive financial rewards or lower electricity rates.
Faruqui and Sergici (2010) found that even with only rudimentary direct load control (air conditioner) capabilities, having enabling technology could increase price elasticity by about one-third. I consider that number a lower bound of what's possible with more sophisticated in-home digital technologies combined with more granular dynamic pricing, with people automating their devices to respond at little inconvenience. This price responsiveness provides more control over electricity bills than in the old analog days with the old analog technologies and old fixed regulated rates.
It also provides demand flexibility that was unimaginable in our analog past. Let's take Marshall's scissors metaphor and apply it to the reality that both demand and supply can and should be flexible in electric systems. The old technological and institutional equilibrium was all supply response and fixed regulated rates. The new equilibrium should be a mix, with the specifics depending on which one can respond at highest marginal value/lowest marginal cost.
I said this was an ode to demand, so I asked ChatGPT to discuss how digital technology and dynamic pricing make demand more flexible, in the style of a Homeric ode:
Sing, Muse, of digital energy's bounteous boon,
Where consumers grasp power, like Gods, to commune.
By automating responses to RTP's call,
They embrace lower prices, mighty and tall.
Oh, consumers, wise and shrewd, take heed,
For RTP unveils its secrets, a sacred creed.
Through technology's touch, a symphony of control,
They dance with the market, a harmonious role.
Lower prices, a treasure, now within reach,
As appliances bend, with wisdom, to teach.
Smart thermostats, guardians of climate's might,
Adjust the temperature, saving energy's light.
Home automation, a grand tapestry it weaves,
As appliances, obedient, serve as leaves.
Lights dimmed, entertainment paused, with grace,
A ballet of efficiency, in this sacred space.
Electric vehicles, charging stations entwined,
With RTP's whispers, their destiny they find.
When prices are low, the charging begins,
A symphony of electrons, a virtuous hymn.
Through automation, they unlock the gate,
To savings and thrifty, a noble state.
But beyond the realm of personal gains,
A sustainable future, the world sustains.
For as consumers heed RTP's decree,
They join the chorus, a grand harmony.
Demand flexes, grid's burden relieved,
A dance of electrons, to our planet's reprieve.
Efficiency grows, as consumption adjusts,
Like ancient rivers, in harmony it thrusts.
A symphony of kilowatts, a virtuous score,
Contributing to a greener world, forevermore.
So let us celebrate, this wondrous age,
Where consumers wield technology's sage.
By automating responses, they claim their might,
Lowering bills, while keeping the future bright.
In Homer's verses, let this ode resound,
To digital energy's triumph, unbound.
A tale of RTP's prowess, by all confessed,
Consumers empowered, their energy blessed.
As a person who lives in a somewhat automated house, the key to successful automation is to make things a) truly automatic, b) understandable, and c) don't try to optimize so much that people feel the need to frequently override the automation. From an energy management perspective quantity automation like thermostats is fairly easy. The time shifting automation mostly isn't, at least without a battery system.
Another personal anecdote: the easiest way to manage energy demand is to have a passively energy efficient house. We moved from an 1890s home last updated just after the start of the 21st century (not much attention paid to efficiency) to one built in 2017. The new house is about 25% larger. The electricity costs are about the same. Modern construction energy efficiency is amazing. ComEd's calculator estimated that total costs for the time of day pricing would be almost exactly the same as the fixed rate plan (as of mid-2022).
When we moved the old house wasn't quite a tear-down candidate and also wasn't quite a gut rehab candidate. Would "we" be better off to apply a larger discount to older housing stock? IMO yes for reasons that include but are not exclusively efficiency related. I realize that's just me.
One typo: "By connecting to the internet and leveraging sensors, weather forecasts, and user preferences, these devices CAM optimize energy use based on occupancy patterns...." I think that should be "can".
Feel free to delete this post.