To continue to reduce the cost of solar energy and other clean energy technologies, scientists and engineers will likely need to focus, at least in part, on improving technology features that aren’t based on hardware, according to MIT researchers. The researchers describe their findings and the mechanisms behind them today in Nature Energy.
Although the cost of installing a solar power system has fallen by more than 99 percent since 1980, this new analysis shows that the “soft technology” features required to deploy a solar power plant, such as structured permitting procedures, supply chain management techniques, and system design processes, contribute only 10 to 15 percent of that total cost decline, with the majority coming from improvements in hardware capabilities.
But as soft technologies increasingly make up the total cost of installing a solar power system, this trend threatens to slow future cost reductions and hinder the global transition to clean energy, said Jessica Tranquil, the study’s lead author and a professor at MIT’s Institute for Data, Systems, and Society (IDSS).
Tranquik’s co-authors include lead author Magdalena M. Clément, a former IDSS graduate student and postdoc who is now an assistant professor at the Hong Kong University of Science and Technology; Goksin Kavlak, a former IDSS graduate student and postdoc who is now a research associate in the Brattle Group; and James McNerney, a former IDSS postdoc who is now a senior research associate at the Harvard Kennedy School.
The team developed a quantitative model to analyze the evolution of the costs of solar energy systems, capturing the contributions of both hardware and soft technology features.
The framework shows that soft technologies have not improved much over time and that the contribution of soft technology features to overall cost savings is even lower than previously estimated.
To reverse this trend and accelerate cost declines, engineers could first look at making solar energy systems less reliant on soft technologies, or they could tackle the problem directly by improving inefficient deployment processes, according to their findings.
“Really understanding where the efficiencies and inefficiencies are and how to address those inefficiencies is crucial to supporting the clean energy transition. We’re putting huge amounts of public money into this, and soft technologies are absolutely essential to making that money work,” Trancik says.
“But until now, we haven’t thought about the design of soft technologies as systematically as we have about hardware. This has to change,” Clement added.
The hard truth about soft costs
The researchers found that the so-called “soft costs” of building a solar power plant (the costs of designing and installing the plant) have become a significantly larger share of the total cost. In fact, the soft cost share now typically ranges from 35 to 64 percent.
“We wanted to take a closer look at where these soft costs are coming from and why they aren’t coming down as quickly as hardware costs over time,” Trancik says.
Until now, scientists have modeled changes in solar energy costs by breaking down the total cost into additive components – hardware and non-hardware – and tracking how these components change over time.
“But if you really want to understand where this rate of change is coming from, you need to dig a step deeper and look at the characteristics of the technology, and things start to diverge in a different way,” Trancik says.
The researchers developed a quantitative approach to model the change in solar energy costs over time by assigning contributions to individual technology features, including both hardware and soft technology features.
For example, their framework captures how much of a decrease in system installation costs (soft costs) is due to standardized methods by certified installers (soft technology feature), and how much of that same soft cost is due to increased efficiency of solar PV modules (hard technology feature).
Using this approach, the researchers found that hardware improvements have the greatest impact on reducing the soft costs of solar PV systems. For example, the efficiency of solar PV modules doubled between 1980 and 2017, reducing overall system costs by 17 percent. However, about 40 percent of the overall cost reduction can be attributed to soft cost reductions associated with improved module efficiency.
The framework shows that hardware technology features tend to improve many cost elements, while soft technology features have only a small impact.
“We see this structural difference even before we collect any data on how technologies have changed over time, which is why mapping the network of technology cost dependencies is a useful first step to identify the drivers of change, not just for solar PV but for other technologies as well,” Clement says.
Static Soft Technology
The researchers used their model to look at multiple countries because soft costs can vary widely around the world: Germany’s soft costs of solar energy are about 50% lower than in the United States, for example.
The analysis found that advances in hardware technology are often shared globally, and their costs have fallen dramatically regardless of location over the past few decades. Innovations in soft technology are not typically shared across borders. Moreover, the team found that countries that were better performers in soft technology 20 years ago are still better performers today, while countries that were worse performers have not seen significant improvement.
Trancik said the differences across countries could be due to market dynamics such as regulations, permitting processes, cultural factors or the way companies interact with each other.
“But not all soft technology variables are things you want to change in the direction of reducing costs, like lowering wages, so there’s more to consider when interpreting these results than just lowering the cost of technology,” she says.
Their analysis suggests two strategies for reducing soft costs: First, scientists should focus on developing hardware improvements that make soft costs more dependent on hard technology variables and less dependent on soft technology variables, such as creating simpler, more standardized equipment that can reduce installation time in the field.
Alternatively, researchers can directly target soft technology functionality without modifying hardware by creating more efficient workflows for system installation or automated authorization platforms.
“In practice, engineers often pursue both approaches, but separating the two in a formal model makes it easier to target innovation efforts by leveraging specific relationships between technical attributes and costs,” Clement says.
“When we think about information processing, we often leave out the very low-tech ways in which people communicate with each other. But thinking about it as a technology is just as important as designing advanced software,” Trancik points out.
In the future, she and her collaborators hope to apply this quantitative model to study the soft costs associated with other technologies, such as electric vehicle charging or nuclear fission. They are also interested in better understanding the limits to improving soft technologies, and how to design better soft technologies from scratch.
This research was funded by the U.S. Department of Energy’s Solar Energy Technologies Office.
