Managing Expectations on Value-Stacking

The discussion around “value-stacking” rates has been cropping up again in the distributed space – primarily in discussions around batteries and microgrids – along with frustration at the lack of movement toward such a structure. While value-stacking is a good paradigm and most electricity markets will get there, in some form eventually, I’ve found this frustration is rooted more in mismanaged expectations than some imagined backroom boogeyman.

What is “value-stacking”?

There are several technical capabilities necessary for a stable electrical grid to move power from the point of generation to the point of consumption.  These include frequency and voltage regulation, instant generation (“black start”) to cover gaps or failures, and spinning reserves. In most markets, ancillary services needs are provided by major generators (IPPs) bid into an ancillary services or capacity market mechanisms in parallel to their energy generation bids. To compete in these markets, a provider must meet a certain capacity criteria (usually at least 100kW) – too big for many distributed energy systems.

Battery Services to Multiple Points on the Grid

Credit: NREL

However, with multiple different types of technology affixed to the grid on the distribution end of the system – many of which can provide at least some if not all of these ancillary services at the point of need – there is an advancing argument that these services should be disaggregated and priced in some way for more technologies and providers to compete.  The NREL diagram to the right illustrates the multiple services batteries can provide at multiple levels of the grid.

The idea of value-stacking is to enable a single site with a single technology, or cluster of technologies acting as single entity, to provide some or all of these services  – receiving revenue for each type of service in the “stack.”  The challenges to implementing a value-stacking system are not insurmountable, but they are complicated and thus their solutions are complex and time-consuming.

Challenge 1: Multi-variable Algorithms and the Point of Diminishing Returns

This challenge assumes a market-based structure establishing the clearing price for each piece of value-stack.  This presents both technical and economic limiters to overcome.  Technically, if you want a competitive market system based on known needs in addition to instantaneous (spot) bid/response/pricing signals when unanticipated ancillary services are needed, you need a software platform that is running multi-variable algorithms in real time, and a circuit system that is adapting to the location of those services as they are provided.  This is technically possible, but it’s not technically present in current RTO/ISO platforms.

In addition to the software challenge, not all ancillary services are created equal.  The value proposition of a single service, like black start, may not be enough to cover the cost of production and deployment of a product that provides.  Hardware, or hardware clusters, would only reach a profitable threshold if they are able to provide multiple services per unit.   However, if a DER provider must have multiple services available to viably recover revenue, why disaggregate services at all?  The answer is time – a cluster of services will not all be needed at the same time at the same price on any given day/month/season.  This leaves system operators trying to determine the minimum cluster of services that must be available from any bidder to feel comfortable that the provider winning a bid for any one service is financially stable enough to still be around to provide it when it is needed.

It also leads to a challenging financing conversation for executives that goes something like “Our widget can provide multiple services to the grid at unknown prices for an unknown period of need through an unknown number of occurrences. Would you like to invest?”

Challenge 2: Market vs Rates

The solution to the first challenge, at least the economics and financing part of it (the technology answer is “consistently write and maintain new software”), is to set a specific price for each service. Forget the open market signals and just theoretically calculate the average value and set a rate.   Every time a participant provides the service, they get paid or credited at that rate – multiple services, multiple payments.  This solves the financing problem for service providers, negates the need for increasingly complex multi-variable algorithms in RTO/ISO platforms, and establishes the point of diminishing returns for investment into providing any one service. This is why most people talk about value-stacking rates – it’s like net metering for more than just generation.

So, what’s the challenge? That it’s like net metering for more than just generation.  While no economic policy is without politics, for the most part, market-based pricing signals based on demand requirements take a lot of the special interests out of day-to-day financial returns.  Establishing a rate, or multiple rates, opens the door of those interests to argue in favor of stakeholders’ priorities ranging from system stability to continued financial viability.  The fight over net metering in states across the country reflects how complicated, convoluted, and in some cases downright ugly, the fight over non-utility owned services can get in rate cases.

While a value-stacked rate structure is the most likely outcome in the long-term to provide the best solution at the lowest customer cost for ancillary services, the fight over that structure hasn’t really even begun. Simplifying the economic solution with set rates for various disaggregated services has the unfortunate consequence of complicating the technical management of the grid itself.

If services are being provided at the distribution level, does the utility determine the type of signal and to whom it is sent on the grid when ancillary services are necessary? How do they manage who “wins” for any particular need? What investments are needed to provide such signals?  How will they get audited? Who is responsible if the service isn’t provided and the grid destabilizes? Most importantly, will utilities be allowed to rate-base the cost of technical platforms and management systems to enable such services even though they aren’t infrastructure capital costs?

Not If, But When

I have yet to interact with any organization, as an employee, a consultant, or a customer, that doesn’t have at least one tale of “great idea, lousy execution.” In the electricity sector, lousy execution can have dire results – economically, technically, or sometimes both.  Regulatory bodies are risk averse, utilities are slow to change, and market models are shackled by inertia. This is not to say that the mechanism for distributed grid service providers will never manifest – keep an eye on Hawaii’s new performance-based revenue structures for early innovations.  But be realistic about the timing when thinking about expected revenue and growth, in any sales and savings pitch, or level of confidence that your side will win in the coming value-stack debates.

Posted in Uncategorized.

Leave a Reply

Your email address will not be published.