Junko Movellan

The Skies are Brightening as Manufacturers Resume Spending to Improve Efficiency

Almost one decade ago, Japanese PV makers dominated global PV production — Sharp (SHCAY), Kyocera (KYO), Sanyo (now part of Panasonic) and Mitsubishi Electric represented about 50 percent of global production in 2005. When German and other European markets expanded quickly, a great number of companies in Europe and Asia, specifically China, jumped into the “potentially” profitable PV industry. They rapidly ramped up their production and brought down costs, leaving Japanese companies behind.

When the Japanese government decided to pump life into the lagging domestic PV market, it created a generous feed-in tariff (FIT) program. Japanese manufacturers began enjoying full access to the lucrative domestic market and started to see the improvements in their bottom lines.

Taking advantage of the uptick in business, Japanese manufacturers have put all of their resources into the domestic market. They have increased shipment and production, improved cost structure, and moved beyond the “module only provider” phase through horizontal and vertical expansion into the downstream solar value chain.

Domestic Market Focus

Japanese manufacturers were export-oriented due to the better profit margin they could earn in German and other European markets. However, that trend is now over. At 1Q’13, Japanese PV makers kept 90 percent of what they produced in the domestic market, compared to just about 30 percent at 1Q’09 (Figure 1).

Figure 1: Japan PV Domestic Production: Exports vs. Domestic Shipment

Japanese solar manufacturers have taken a “Japan shift,” said Nobuyuki Nakajima, Solar Frontier’s manager of communications.  A few years back, Solar Frontier was export-focused, but since 2012 its domestic shipments have exceeded exports — about 80 percent of its modules will serve the growing domestic market in 2013, explained Nakajima.

Solar Frontier, a CIS (copper, indium, selenium) thin-film PV manufacturer, made its first-ever operating profit in the first quarter of 2013, two quarters ahead of plan. The company successfully reduced its material costs by 25 percent through the first half of 2012, bumped up its production capacity utilization to 100 percent at its 900-megawatt (MW) Kunitomi plant in January, and will resume production at its previously suspended 60-MW Miyazaki No. 2 PV plant in July to keep up with demand.

Kyocera, a vertically integrated poly-crystalline silicon PV manufacturer, has also been improving sales and profits by shifting its focus largely to Japan. According to Ichiro Ikeda, Kyocera’s general manager of solar energy marketing division, its domestic shipment accounted for about 80 percent of the company’s global shipment in FY2013 (April 2012 – March 2013), compared to about 50 percent in FY2010. It has been meeting the growing domestic demand by re-importing modules from its overseas production facilities in Czech Republic, China and Mexico. Ikeda said that Kyocera is planning to boost its shipment to over 1 gigawatt (GW) for this fiscal year (April 2013 – March 2014), up from 800 MW in FY2013.

PV Module Technology

The PV technology mix in Japan has also been changing. Domestic manufacturers largely produced poly-crystalline silicon (poly-si) technology, so it dominated the market. However, the Net FIT for the residential market revitalized the domestic market. Since then, the demand for high-efficiency or mono-crystalline silicon (mono-si) modules has gained popularity among homeowners who want to maximize energy production on their space-limited roofs.

SunPower (SPWR) and Panasonic, providers of world-leading, high-efficiency modules, are currently neck and neck, chasing the largest market share in the residential segment in Japan. Sharp and Mitsubishi Electric, previously poly-si module focused producers, started offering mono-si modules specifically for residential customers. Last year, Mitsubishi Electric announced the termination of poly-si modules production to focus on mono-si sales. Sharp, Japan’s largest PV producer, but deeply financially troubled, announced its very first outsourcing contract deal with SunPower to sell SunPower’s high efficiency modules under Sharp’s brand (“Black Solar”) for the domestic residential segment.

Although demand for mono-si modules is expected to grow, the launch of the full FIT program has ignited the large-scale, non-residential system market. Since its launch, the demand on more price-competitive poly-si modules has started to pick up again.

Sharp’s sales manager stated that Sharp’s current tactic is to ship mono-si modules to the efficiency-focused residential segment, and ship poly-si to the cost-conscious, non-residential segment.

The residential segment has been bread-and-butter for the Japanese module makers, providing a steady market with good profit margins; however, the module makers cannot ignore the potential growth of the non-residential segment, which is expected to grow much larger by volume than the residential segment in the next few years.

According to the Japan Photovoltaic Energy Association (JPEA), the non-residential segment grew by close to 900 percent in FY2012 (April 2012 – March 2013) from FY2011 (April 2011 – March 2012) while the residential segment grew by 55 percent. For the first-time ever, the non-residential segment exceeded the size of the residential segment.

Panasonic, a long-time producer of HIT (premium, high-efficiency modules) has even started offering OEM poly-si modules to capture the piece of the growing non-residential segment. In April, the company shipped 8,784 240-W poly-si modules, 2 MW in capacity, to a FIT non-residential project in Tokushima Prefecture – its biggest poly-si project in Japan.

The thin-film market is also making headway in Japan against silicon counterparts. A recent report states that thin-film PV lost more ground globally to silicon PV in 2012; however, the thin-film share in Japan is, in fact, increasing (Figure 2).  Data released by JPEA shows that thin-film took 21 percent of Japanese PV technology market share in Q4’12, up from a 4 percent share in Q1’10. Solar Frontier is the biggest contributor to the growth of this segment.

Figure 2: Japan Domestic PV Market by Module Technology

Vertical and Horizontal Expansion

To protect its turf and profitability, Japan PV manufacturers are expanding their product and service offerings and strengthening their domestic networks against foreign PV markets, which now accounts for more than 30 percent of the domestic market.

Kyocera, Sharp and Solar Frontier have moved beyond “module only provider,” by vertically expanding into the downstream solar value chain, as an EPC contractor, project developer and independent power producer.

Solar Frontier has created an investment company, SF Solar Power, with the Development Bank of Japan (DBJ) to fund around 100 MW worth of medium-scale PV in Japan. These projects serve as a “sweet spot” since they are easier to acquire and interconnect than projects over 2 MW.

Last year, Kyocera joined forces with IHI Corp. and Mizuho Corporate Bank to construct one of a 70-MW PV project, the nation’s largest, in Kagoshima Prefecture. Kyocera will not only supply its modules but also undertake part of its construction, operation, and maintenance. The project is expected to be completed by this fall.

In terms of the horizontal integration, Kyocera, Sharp and Panasonic all have starting selling lithium-ion storage batteries with PV systems for the residential segment in order to offer the complete packaged solution to “create, store and control energy.” Kyocera will also add Home Energy Management System (HEMS) to its PV and lithium-ion battery system offerings.

To the World

The Fukushima Disasters in March 2011 certainly created a keen interest and demand for safe and clean renewable energy sources, including solar, but a solar revitalization plan had already been in the works.

In 2010, JPEA released “JPEA PV Outlook 2030,” which spells out JPEA’s vision to create ¥10-trillion (about $100 billion) Japanese PV industry and increase the share of Japanese PV makers or “Japan Brand” to 33 percent of the world PV supply by 2030, up from 8.5 percent in 2011. “Japan Brand” means modules marketed and produced by Japanese companies not only in Japan but also in other parts of world.

According to the Outlook, the domestic market will be saturated by 2020.

After that period, the survival of Japanese PV manufacturers will depend on how much they can expand outside the domestic market. The current revitalization of the domestic market is providing them with a chance to regain the strength required — technology, innovation, and production capacity — to last in this turbulent PV industry.

Junko Movellan is a Solar Industry journalist who writes and analyzes the US and Japan PV downstream markets. She has more than 10 years of experience in the PV industry, analyzing and developing business strategies for global companies. She previously worked as a Senior Analyst at Solarbuzz and as a Market Development Analyst at Kyocera. She is based in California, USA.

This article was first published on RenewableEnergyWorld.com, and is reprinted with permission.

Tom Konrad CFA

This article was first published on the author’s Forbes.com blog, Green Stocks on May 31st as “Kandi Technologies: Weighing The Evidence.” I have since added a short update to the end of the article.

Last year, I brought Chinese off-road vehicle and electric vehicle (EV) manufacturer Kandi Technologies (NASD:KNDI) to readers’ attention.  I like Kandi because the company was already profitable and trades for a tiny fraction of what a US-based EV maker would.

The Strategy

I also like Kandi’s electric vehicle strategy, which focuses on inexpensive commuter vehicles combined with battery-swapping.  While this sounds a lot like the the strategy of recently bankrupt Better Place, Kandi’s strategy avoids one of the biggest problems with Better Place’s strategy: Kandi does not have to bear the expense of extra sets of batteries or swapping infrastructure.  The batteries are owned by the local utility, which can use them when they are not in cars to provide stabilization and ancilliary services to the grid.  Kandi just (profitably) manufactures the cars and licenses the battery swapping IP.

With the Chinese government in Beijing pushing hard for more “New Energy Vehicles” as the Chinese call EVs in order to cope with its horrific problem of urban pollution, even China’s largest privately owned automaker, Geely (HKEx: 175, OTC:GELYF) got religion, and has signed a 50-50 joint venture to produce EVs with Kandi.

The Valuation

With Kandi already profitable based on its legacy ATV business, I and other Kandi shareholders have long been frustrated that Kandi does not trade at a much higher multiple of earnings and revenue.  Kandi has a trailing P/E ratio of 20, and trades at less than 2 time trailing revenue.  Meanwhile EV high-flyer Tesla (NASD:TSLA) is trading at 13.5 time revenue, and 100 times next year’s expected earnings (Tesla lost money last year, and is expected to only break even in 2013.)

The China Price

There are several factors contributing to Kandi’s low valuation:

§ Kandi got its Nasdaq listing through a reverse merger, a strategy which was followed by a number of other Chinese companies, many of which were later found to have fiddled their books, absconded with shareholder funds, or otherwise been frauds.

§ As a result of its small market capitalization and the general wariness of Chinese stocks, no analysts follow Kandi.

§ A number of negative articles, many of which were written by investors who were short the stock, have highlighted irregularities in Kandi’s listing process and past reporting.

Although the negative articles about Kandi have been disturbing, none of them have turned up anything wrong with Kandi’s  financial accounting.  I’ve generally taken this as a good sign.  When the same group of people who made good profits by shorting Chinese stocks and then exposing their accounting frauds have been unable to turn up anything so serious about Kandi, I have to wonder if there is anything to find.

With this in mind, I set out last month to parse through the novel-length and rather dense Kandi-bashing articles to demonstrate that there was nothing there for investors to worry about.  I failed, and instead found myself doubting the judgement and/or honesty of Kandi’s management.  I’d like to emphasize that there is no proof of wrongdoing, but investors who wish to hold on to their money don’t have the luxury of waiting until their suspicions are confirmed beyond a reasonable doubt.

Sharesleuth

The most in-depth articles looking into irregularities surrounding Kandi were written by Chris Cary of Sharesleuth.  Carey is a former reporter for theSt. Louis Post-Dispatch.  Sharesleuth is funded by Mark Cuban, who often trades on the information Carey digs up before he publishes his articles.  Some people find this business model distasteful  but as Carey puts it, “If Sharesleuth.com exposes fraudulent companies and Mark Cuban uses profits from trades to finance more investigative reporting, then I’m OK with that.”  I’m also OK with it.  I don’t see the difference between Sharesleuth and any mutual fund manager who goes on CNN to talk about his portfolio.  Or between Sharesleuth and a blogger writing about stocks he owns on Forbes orSeeking Alpha, for that matter.

The controversy about Sharesleuth’s business model mostly seems to arise because most of Cuban’s trades are on the short side.  But in the case of Kandi, Cuban was never short.  I asked Carey about this in an interview, and he responded that Kandi is a very difficult stock to short.  For a billionaire like Cuban, the money he could make shorting a tiny stock like Kandi is hardly enough to move the needle.  Carey uncovered the information in his articles in the course of investigations into the people who brought it public, along with ten other Chinese reverse merger companies, including New Oriental Energy (OTC:NOEC), Telestone Technologies Corp. (Nasdaq: TSTC); and Orsus Xelent Technologies Inc. (OTC: ORSX).  Many of these have since been delisted, and Kandi is virtually alone among them for not trading well below its initial offering price.

Clearly, Kandi should not be indited based on guilt by association, and the scrutiny the company has been under because of these associations should give us some confidence that any past misdeeds are either very well buried or have already been revealed.  Nor do any of those misdeeds reach the level of the outright accounting fraud found in many of the Kandi’s reverse-merger brethren.  Kandi has not been accused of anything illegal.

My distillation of the Sharesleuth revelations is:

§ A number of people made millions off Kandi’s reverse merger, and these people were never properly identified in the company’s SEC filings.

§ From 2009 to 2011, Kandi significantly overstated the number of EVs it sold.  After Sharesleuth showed that Kandi’s claimed sales of EVs were not supported by the number imported or sold by Kandi’s dealers, the company quietly revised its financial statements, revealing that many of its claimed EV sales were actually sales of gas powered vehicles.

The Defense

 

Kandi’s defenders dismiss the first point as old news, saying that what should really matter to investors is Kandi’s current prospects.   To the second point, they say that Kandi has admitted its mistake, and the miscategorization of sales of gas powered cars as EVs made no difference to Kandi’s revenue or earnings in any of the affected years.

They also emphasize that Kandi is not accused of any criminal act or fraud, and attempt to undermine the credibility of Kandi’s detractors by calling the negative articles paid hit pieces.  Of course, Kandi’s defenders are long the stock (as am I), which is at least as much of a bias as being short.

My Take

I’m certainly happy that Kandi has not been accused of fraud, and I do prefer to focus on Kandi’s future than on events which occurred before I was ever a shareholder.  On the other hand, when we’re trying to predict how management will behave in the future, our best evidence is how they have behaved in the past.

In the case of unknown individuals profiting from the reverse merger, this was at best bad judgment on the part of Mr. Hu, Kandi’s President, CEO, and largest shareholder.  The reverse merger process seems to have needlessly diluted existing shareholders, and also shows Mr. Hu working with a number of unsavory characters, perhaps unwittingly.  At worst, Mr. Hu and his associates may have directly benefited from the transactions in ways which were not disclosed.

Either way, the incident undermines my faith that Mr. Hu will do everything in his power to protect the equity of the company’s current small shareholders.

In terms of the misreported EV sales, the best case scenario is that it was simply a translation mistake.  I find this scenario unlikely, because the exaggeration occurred repeatedly over a couple years.  Nor does the fact that the mis-categorization of EV sales did not affect reported sales or revenues mean that the number of Kandi’s EV sales was not material to investors’ investment decisions.  The Kandi “story” depended on the growth of its EV business even then: Here’s an article from 2010 making the link explicit: Kandi Tech Reports Strong Results, But Future Depends on Electric Car Growth.

Other Evidence

After  couple of my picks recently revealed that they would have to restate their financial accounts because of misreported revenue, I began using theBeneish M-Score as an early warning system for earnings manipulation.  I calculated Kandi’s M-Score based on annual accounts from 2010 to 2012, and on quarterly accounts for the last three quarters.  The M-Score combines factors which might give a company an incentive to manipulate with factors which pick up the distortions caused by common forms of earnings manipulation.  Details about how to calculate M-Score and a spreadsheet can be found here. For nearly all the periods I tested, M-Score indicates that Kandi has a moderate chance of having performed some earnings manipulation.  Exactly what this probability is is hard to say, but the M-Scores are a long way from giving an “all clear.”.  The 2010 annual report looks most likely to have been manipulated, mainly because of a high level of receivables growth relative to sales growth.  Note that this period coincides with the inflated EV sales numbers.

Companies can have high M-Scores without having manipulated earnings, but a high M-Score says “proceed with caution.”  Maxwell Technologies (NASD:MXWL) had an M-Score in the third quarter of 2012 that was similar to Kandi’s annual 2010 M-Score, and the next quarter they announced that they had been misreporting revenue since 2011.  (I suspect Maxwell’s mis-reporting may be greater in extent than has yet been revealed.)  M-Score will not flag all earnings manipulation, but it may flag some honest companies as well.

Reading through Kandi’s filings, I noticed that Kandi’s largest shareholder at the time of its listing was ExcelVantage Group, a fund controlled by a Chinese retiree Tim Ho Man.  In 2010, Mr. Tim transferred control of ExcelVantage to Kandi’s CEO, Mr. Hu, “pursuant to a Transfer of Equity Agreement” between them.  Kandi’s listing documents made no mention of any connection between Mr. Hu and Mr. Tim.  While it is possible that Mr. Hu bought ExcelVantage from Mr. Tim in an arms-length transaction, it seems more likely to me that the two men had an undisclosed agreement between them which gave Mr. Hu effective control of ExcelVantage all along.

Once again, there is nothing illegal about this, but it had the effect of obscuring the fact that Mr. Hu retained a controlling stake of Kandi at the time it went public.  That’s something I would have wanted to know had I been considering investing at the time.

Conclusion

There are a number of instances and red flags about Kandi’s management that lead me to want to proceed with caution.  At the very least, the company has not been forthcoming with relevant information that investors would have been interested in.  A company looking to build a reputation for good shareholder relations would have disclosed this information.  At worst, the company may have intentionally misled investors regarding its EV sales at a time when its accounts also showed signs of possible distortion.  If that’s the case, it would be reasonable to assume that they will do something similar in the future.

On the other hand, the evidence of Kandi’s current progress at building acceptance for its EVs is based not only on the company’s statements, but a large number of articles in the Chinese press, and agreements with Geely and a number of Chinese cities and provinces.  My feeling from this is that Kandi will continue to rack up good press and increasing EV sales for the rest of the year.   The fact that Kandi also recently filed an S-3 to allow it to sell additional securities also leads me to believe that, if the company is likely to exaggerate its results, it will do so in the coming months in order to boost the share price.

If you would like to read the full bull case for the stock, the best place to start is to read these three recent articles by Art Porcari.

After weighing the evidence, I no longer consider Kandi a long-term hold.  That said, my concerns about management are long-term in nature, and I think Kandi’s short term trend will be up.  This article itself may cause a downward blip, but Kandi’s shareholders are so used to negative articles about the stock that I doubt this one will have any long term effect, and I expect Kandi’s upward momentum will soon resume.  I intend to maintain my reducedholding to take advantage of that trend.

Update 6/10/13: The upward climb I predicted above started much sooner than I thought, when a relatively minor news story about a car developed for the Kandi-Geely JV received approval from the Chinese government, and Kandi finally caught some of the Tesla (NASD:TSLA) fever.  When I wrote this article 10 days ago, Kandi was trading around $3.80, today it’s trading at $6.50.  I’m now mostly out of the stock, having sold covered calls at $5, but I’m not ready to call a top.  As Tesla showed, once an EV stock catches investors’ imaginations, it can completely defy gravity and fundamentals.  Bottom Line: If you still own Kandi, enjoy the ride, but this hot EV stock should be a rental, not a purchase or even a long term lease.

Disclosure: Long KNDI stock, short KNDI covered calls, MXWL.

DISCLAIMER: Past performance is not a guarantee or a reliable indicator of future results.  This article contains the current opinions of the author and such opinions are subject to change without notice.  This article has been distributed for informational purposes only. Forecasts, estimates, and certain information contained herein should not be considered as investment advice or a recommendation of any particular security, strategy or investment product.  Information contained herein has been obtained from sources believed to be reliable, but not guaranteed.