Happy New Year – Let the Confusion Begin

This is the time of year when a lot of counting and announcing goes on - and much of it can be very confusing. In general, 2009 will be a confounding and frustrating year to parse — cheap cell and module prices all year long led to a lot of cross buying and the 2-gigawatts of inventory at the beginning of the year doesn’t help un-confuse things. Add to this the usual flood of announcements (primarily PR), and the fact that not everyone counts the same thing, and the PV world can be a pretty confusing place to do business right now.

A correct count of shipments, production, capacity and installations is important (and these are all different numbers) — all of these business metrics help manufacturers and demand side participants do their business planning. Even with all involved participating in the noise of various numbers and announcements, an accurate count is crucial. In the past few years it has become a game of who do you believe — noise aside, sometimes it is difficult to tell the announcement that will come true from pure (and well meaning) public relations — and all numbers mean something to someone for often quite different reasons.

Lobbyists, visionaries and public relation professionals and sometimes governments like high, optimistic numbers because these numbers make a point — often this point is that the sky (pardon the pun) is the limit for solar industry growth. For manufacturers accurate leaning towards conservative numbers are crucial for business planning. For startups, optimistic numbers are important for attracting investors. To some groups, often governments, sometimes utilities, and always journalists, installations are important — what goes into the ground (or on the roof) in a given year tells them the amount of solar electricity that is currently being generated. Many people think shipments, production and installations are the same number.

In any case, someone somewhere will always be counting something either by primary research, by gathering information from various secondary sources and compiling it, or sometimes by simply guessing.

So… various numbers for 2009 will abound this year… this is a primer on some things to keep in mind.

1. Shipments are delivered sales to the first point of sale in the market — production and shipments are different numbers — always have been, always will be.
2. Production is what a facility produced in a given calendar year — that may or may not have left the factory to become a shipment.
3. Commercial capacity is what a facility is capable of producing and then shipping in a calendar year.
4. Announced capacity increases are what a manufacturer SAYS they will add in capacity and that may or may not be added depending on business conditions.
5. Nameplate capacity is what a facility is capable of producing in a perfect world.
6. Installations are what actually went into the ground, on the roof, or was built into a building.
7. 2009 started out with ~2-gigawatts of inventory, primarily on the demand side. This inventory was sold, resold and resold again and eventually installed — so, installations will be higher than shipments.
8. Prices plummeted in 2009 due to high inventory levels and sluggish demand for most of the year. This led to high levels of outsourcing — manufacturers who typically assemble their own cells into modules bought cheap (often below cost) cells from other sources and assembled those. Then, everyone claimed the outsourced cells as production. So, shipment and production numbers will range wildly and finding the source of the technology will be quite an investigative chore. Result: 2009 estimates for production and shipments when all pulled into the same chart will look like a craggy mountain range.
9. Remember that sales backlogs are not shipments — counting what actually left the factory is an important business metric, and also useful in assessing installations (and forecasting installations).
10. For every forecast, ask what is being counted and how it is being counted — a forecast based on different analysts’ assessments (given that they may all be counting different things) is not useful.

…

And, if this is not confusing enough, Figure 1 offers a glimpse of the various metrics by which the 2009 year in solar will be assessed — inventory at the beginning of 2010, technology shipments to the first point of sale, production, installations, announced production, commercial capacity and announced capacity — all based on global numbers. (Before anyone asks, I will be reporting shipments to the first point of sale, production, installations, commercial capacity and 2009 inventory brought into 2010.)