With the last of the year's big solar trade events in Europe and the US now out of the way, thoughts are beginning to turn to what next year holds in store. Mark Osborne, PV-Tech’s senior news editor, took the opportunity to speak with NPD Solarbuzz team leader Finlay Colville regarding the level of end-market demand that can be expected in 2015.
Finlay Colville, VP of NPD Solarbuzz updates market forecast for 2015.
With very strong demand expected in China during the final quarter of 2014, where do we stand now on 2014 deployment levels?
Compared to 12 months ago, the dynamics of the industry have held up pretty much as expected. In December 2013, we called the 2014 market at 49-50GW at a time when most others were being highly cautious and forecasting figures at or around the 40GW level.
By the time the numbers were finally counted for 2013, the final consensus was just under 39GW. This triggered a complete change in our forecasting methodology going into 2014, as simply adding up individual countries was not providing the correct means of assessing how the solar industry had changed and was the key factor for why so many market forecasts had been low in the past.
Most of the tweaks on 2014 have been at the decimal-place level during the year, with 49-50GW still our most-likely figure once all the counting is done.
What changes in the industry have caused the demand methodology to be revised?
Solar PV deployment is now being driven by several factors, and simply doing a spreadsheet summation across 10, 20 or 100 countries in isolation only gets you half way there in forecasting deployment levels.
Equally important are domestic manufacturing jobs, and the ability of certain countries to turn on and off end-market growth levels to maintain state or regional employment figures. Also, there is the added impact of sales teams now active on a global scale.
There is enough slack, and enough traction globally at current system ASP levels, to shift supply across countries or regions to fulfil the ever growing pipeline of opportunities. Furthermore, the market growth in recent years has been largely due to the popularity of large-scale, or utility-based, solar farms. In contrast to rooftops, 5-10MW can be added now very quickly by seasoned EPCs assuming land access and grid connections are in place. Many sit with active pipelines of several hundred megawatts.
So, the forecasting models we are using now pull in supply-driven levels, global pipeline activity and, of course, the policy-influenced issues at the country level. The net effect right now is providing an upside delta that may appear to many as somewhat at odds with simply adding a few countries and fudging a rest-of-the-world contribution on top of this.
How is 2015 shaping up, based on this new forecasting methodology?
Again, going back to December 2013 when we changed the forecasting methodology, the figures were coming out at 60.2GW for 2015. Given that many were still thinking 2013 was going to end up a market in the low 30s, it was somewhat a leap of faith to call out a figure as high as 60GW.
Today, we are forecasting 62GW as the most likely end-market demand in 2015. Of course, as with all future projections, there is upside and downside, but the underlying takeaway is of an end-market in 2015 that should hit the 60GW mark, representing 20% annual growth compared to this year.
NPD Solarbuzz expects around 20% growth in 2015.
How does this compare to shipment guidance being issued now from the top tier suppliers to the industry?
Almost every one of the leading module suppliers is preparing for shipment growth, going into 2015. Between them, the top two module suppliers, Yingli Green Energy and Trina Solar, could easily ship 10GW next year. Add to this the double-digit growth forecasts from the other main suppliers, and the figures start to add up quickly.
Some suppliers are still guiding low for 2015, at the low-50s level. Whether this is to set market expectations low, or to cite higher market-share levels, is not entirely clear. It could also be that the fear of committing any new capex is also supported by setting the bar lower going into quarterly reporting discussions.
Like this year, 2015 looks like being dominated by China, Japan and US. What level of market share do you expect from these big three?
Next year, market demand will again be dominated by China, Japan and the US. Europe may stabilise but strong growth there is still some way off. We could easily see a situation in 2015 where two-thirds of new PV installed comes from China, Japan and the US. Other emerging regions will grow for sure, but to reach the 60GW level, it is really what happens in the three main countries that will shape the year as a whole. From a forecasting perspective, it is wiser to look at the collective pull from the big three, rather than the individual levels. This is similar to how things panned out during 2014.
Lastly, will a 60GW market in 2015 finally see an uptick in capex, after several years of downturn?
Well, this is actually still in the balance. For some suppliers, such as SunPower and First Solar, so long as project development is done with in-house technologies, there is no avoiding capex additions, whether greenfield fab build or extensive line upgrades.
But for most of the others, there is still a desire to remain asset-lite and to acquire existing c-Si capacity. A few years ago, the shakeout was expected to see lower tier capacity exit the industry. But this has not happened, and effective capacity levels have increased to above 50GW today.
Announcements are coming fast and furious now, with talk of module assembly now the flavour of the day. So, the issue could come back to cell capacity, and how far this can go in getting supply in place to ensure 60GW of shipments next year.
The safe bet in the past was for cell capacity to be added quickly in China and Taiwan, but trade issues have rather changed this outlook today. For the Chinese, the risk is not as bad as some observers think however because the two biggest markets in the world now, China and Japan, have no obstacles to modules being supplied with China-made cells or wafers. For Taiwan cell makers, it is less clear, and Taiwan cell additions had been widely expected in the second half of the year as a result of the 2012 US trade ruling.