Outsourcing has been a common practice in the photovoltaic industry since…always. Ignoring it in favor of reporting higher shipment numbers has been a common practice since…always. There is more outsourcing now than there was ten years ago because the industry is bigger. When the PV industry was at megawatt levels, outsourcing was at megawatt levels. Now that the industry is at gigawatt levels, outsourcing is at gigawatt levels. Today’s outsourcing is also more acceptable — in the past everyone did it quietly, today it is out in the open. Yet despite this openness and acceptability, double counting continues and the industry continues to be oversized.
Figure 1 presents the various metrics that make up the PV industry during a calendar year with the exception of grid connections. These metrics are supply and demand inventory, commercial capacity, production, shipments, installations and defective modules. Think of production and shipments as conducting a little dance in and around inventory levels. That is, production can be lower than shipments depending on inventory levels. Production can also be misreported, that is, shipments are sometimes used as a stand-in for production and vice versa. In breaking industry activity down into its various components (except for grid connections) it appears as if should be easy to correctly size the PV industry's annual activities. Nope.
Figure 1: PV Industry Metrics, 2013
The Shipment Counting Shell Game
A shell game is a now-you-see-it-now-you-don’t trick of evasive maneuvering. The rules (such as they are) consist of this: there are three cups and one small object such as a stone. The stone is placed underneath the cups, the cups are shifted around and everyone places bets as to where the stone will end up.
Shipments of PV cells/modules are counted in order to arrive at the correct business size of the industry during a specific calendar year. The point is to accurately size the megawatts based on the technology developed and manufactured at point A followed to its arrival at point B; everything from point B on is a new count.
Most manufacturers buy from other manufacturers and many include what they buy as their own production or in their shipment count. Manufacturers with wafer capacity and module assembly capacity send out their wafers for tolling and then assemble the cells in modules in-house. For example, manufacturer A ships 1-GWp of wafers to Manufacturer B who returns 1-GWp of cells to Manufacturer A. Both manufacturers report shipments of 1-GWp of PV cell technology and the PV industry is oversized by 1-GWp.
Another way in which shipment numbers are inflated is when subsidiary relationships are unclear and nontransparent. In this case, Manufacturer A ships 100-MWp to a wholly owned subsidiary that may or may not install the technology and may even ship it back to the parent. In this way and over the course of many such exchanges the PV industry can also be oversized, significantly.
Why This Is Dangerous for PV
The PV industry has been experiencing accelerated growth for quite a few years — not always profitably. It has also been underutilizing its available commercial capacity for quite a few years. In 2013, capacity utilization for the PV industry was 82 percent — based on shipments to the first point of sale. This is a vast improvement over the past few years during which capacity utilization fell at times below 60 percent.
Overtime, the systematic oversizing of PV industry output, whether through outsourcing or through shipping to subsidiaries, has made the industry look significantly more successful than it is and helped (along with too low prices) bring about the end of most of its incentives. Conventional energy, of course, does not need to worry about its success interfering with its ongoing incentives and subsidies.
Pragmatically, stakeholders all along the PV value chain have a vested interest in having access to data about capacity, production, shipments and inventory that is as accurate as possible — for business planning purposes. Successful strategies require good data. Unfortunately, shipment reporting has often been a matter of saving face and looking successful more than arriving at an understanding of what is really happening.
Think of it this way: you are wandering through a shopping mall and you come across a map of the facilities with a helpful icon indicating where you are in terms of the other stores as well as your destination. The helpful icon typically reads: You are here. So, now that you know where you are, you can figure out how to get to where you are going, or, you can decide to go someplace else entirely. If the helpful icon pretended you were further along, it would not be very helpful. The point of shipment and production reporting is to offer accurate information as to where the industry is at a certain point in its history so that it can develop a strategy to get where it wants to go.
The PV Industry Is Here
Figure 2 offers capacity, production shipments and inventory from 2008 through 2013. During this period PV industry capacity to produce commercial c-Si and thin film modules increased by a compound annual 34 percent with production and shipments increasing by a compound annual 44 percent.
Figure 3 presents technology (c-Si and thin film) shipment market shares for 2013. Crystalline technologies had a 91 percent share of shipments in 2013.
Figure 3: Crystalline and Thin Film Shipment Shares 2013
You Are Here, Where Do You Want To Go?
One problem with outsourcing is that the farther production gets from the original manufacturer the lower the quality gets over time and the bigger an industry gets, the more outsourcing is conducted. Currently there are module quality complaints from Japan, the US, Europe as well as other countries. In some cases, quality complaints go back seven or eight years (about the time the PV industry surged into gigawatt levels of shipments). In many cases cells and modules are reshipped so many times that it is almost impossible to pinpoint where the lower standard production began in the first place. Unfortunately, more than one region is responsible, so, pointing fingers is counterproductive to solving the problem.
Outsourcing is not going away and, aside from the unfortunate oversizing of the industry, the focus should be on quality. Back to oversizing the industry, at this vulnerable stage in its always vulnerable history, the PV industry should exert tight controls on both quality and representations of its size so that it can have a clear eyed vantage point from which to assess where it wants to go and how to get there.