Page 10 - North American Clean Energy March/April 2019 Issue

solar energy
Forecasting EPC Prices
Challenges and possibilities
by Bill Hargett
As the Investment Tax Credit step down becomes more imminent in 2020, and beyond, EPCs and installers have seen renewed price pressure on lower-cost delivery of solar systems. As
a matter of course, construction and delivery rms regularly receive RFPs from owners and developers regarding their project pipelines “post-ITC.” Unfortunately, guidance for EPC pricing 4-5 years in the future is extremely unreliable, and often exacerbated as developer- owners compete on PPAs based on “guesstimate” forward cost curves.
SEIA, NREL, Wood McKenzie, and others continue to publish quarterly reports
on residential, commercial, and utility scale system pricing. Even a cursory look at these reports will show that a) yes, pricing has come down 25 percent since 2015, and b) the vast majority of gains are due to module and inverter price dynamics. While installation companies can optimize purchasing through volume orders or master supply agreements, most contractors and developers have similar price visibility into the modules, inverters, and racking prices. ese major components can make up over 55 percent of the turnkey cost, and dramatically change project economics because of prevailing duty and tari legislation. From an EPC perspective, owners may then apply considerable pressure to optimize price on the 40-45 percent balance of plant, which largely consists of direct labor, engineering, and overhead/margin.
In the emerging solar industry of 5-10 years ago, technology and delivery risk incurred higher margins and contingencies, typically in the mid to high teens. Margin expectations have declined signi cantly since the early 2010s (to the mid and low single digits), even as improvements have been made to reduce the amount of direct management and overhead. However, many in the solar industry regularly hear of projects and contractors burned by unexpected geotechnical, labor, permitting issues,
et al. If these risks are not su ciently captured, then they are priced in as contingency or change orders, provided the EPC contract entertains them. Margins are unlikely to be up for negotiation going forward.
As an overall portion of pricing, engineering is a relatively a small percentage (roughly 2-5 percent), though a vitally critical component to the overall project success. While there may be opportunities to o shore some of the drafting work, the nature of local code compliance and local familiarity can often require specialist, time-intensive participation. Combined with a constrained technical-labor market, third-party engineering prices also appear to be increasing.
is leaves labor and productivity improvements. Direct labor can contribute 12 to 15 percent of project costs, with several large caveats: What are the prevailing
and union or open shop wage requirements? Are there competing industries and projects? In short, labor costs and productivity must be measured in terms of the local competitive dynamics, such as how to compete with local projects that reduce labor availability, and how to reduce workforce turnover in the face of challenging site conditions (who wants to work in central Florida or central Texas in the middle of August?). Additionally, EPCs must closely evaluate remote site locations as to how they can nd capable workforces to complete the work.
It’s important to note that installation rates have improved signi cantly over the
last several years; some tracker manufacturers now have a global supply network with the ability to support installation velocities of 10MW to 15MW (or more) installed per week. Much of this improvement has been due to the very signi cant engineering and design work that racking companies have put into reducing the number connection nodes on their equipment, in addition to providing extensive support in optimizing phased delivery (other rms have attempted to make headway in prefabricated racking “tables” that can be e ectively craned/winched from the back of a truck and installed as blocks, though it remains unclear how truly cost e ective this will be given upfront costs of prefabrication). Anecdotally, however, dramatic increases in delivery and productivity rates are unlikely to improve without some form of automation, which has yet to gain industry-wide traction. e incremental process of improving
10
MARCH•APRIL2019 /// www.nacleanenergy.com