LDK Solar: Understanding Recent Financing Activities

by: Investing Hobo

Since last November, when LDK Solar (NYSE: LDK) reported its third quarter 2010 earnings, the company has announced a series of significant business developments. Most of these announcements are linked to LDK’s ongoing debt restructuring, especially ahead of the company's 400m convertible bond debt potentially becoming due in April. In addition, LDK also raised its fourth quarter 2010 guidance.

In a first attempt to revolve its 400m convertible bond debt, LDK announced an exchange offer to extend the due date by two years. As an incentive, current bond holders were offered a one time cash bonus up to 8.5% of the bond value held. Two weeks later, LDK increased the cash bonus to 10%. However when the final results were announced, only 32m of the company’s 300m exchange offers were tendered.

LDK wasn’t alone, since Evergreen Solar (NASDAQ: ESLR) also experienced limited success with its debt exchange offer which closed last month. Of the 265m of ESLR’s debt offered for exchange, only 22.7m were tendered. Clearly, investors had become much more cautious over handing out new capital to what was once a hot sector.

A month after LDK’s failed exchange offer, the company announced a secondary offering. When the deal closed, LDK raised a net 164.2m through the issuance of 13.8m new ADS at 12.40 per share. Although the closing price for this secondary offering is well above LDK’s 52 week low of 4.97 last year, it represents a steep discount to the 39.29 conversion price on the company's convertible bonds. Had LDK’s exchange offer succeeded, significant dilution could have been avoided-- assuming the company’s shares returned to pre-2008 credit crisis levels.

The secondary also reduced Mr. Peng’s (LDK’s chairman and chief executive) ownership position to 50.1%. It’s not unreasonable to conclude the maximum size for this secondary was limited such that the company’s founder maintained majority ownership. If Mr. Peng’s intention was to retain majority ownership, future capital raisings may have to exclude dilutive deals, such as additional secondary offerings.

Less than two weeks after the closing of LDK’s recent secondary, the company announced yet another financial offering. With dilutive deals potentially off the table in order to maintain the CEO’s majority ownership position, LDK announced a 1.2b RMB (181m USD) bond offering, bearing a staggering 10% annual interest rate. The high interest rate offered was surprising because much of the company’s existing debt averaged just 5-6% annual interest. Why would LDK have to offer such high interest rates to raise capital when months earlier the company announced a massive 60b RMB (8.9b USD) credit facility with China Development Bank (“CDB”)?

Contrary to the views of some solar industry pundits, China is not subsidizing their solar industry with free or low interest loans. Chinese banks offer credit to Chinese solar companies with the same guidelines as for any other industry. Loans and credit lines have restrictions and covenants according to annual reports filed by major Chinese solar companies including LDK. Interest rates have ranged from 5% to as much as 8%, depending on financial ratios. The interest rates offered by Chinese banks for their credit facilities are in fact generally higher than interest rates many solar companies have been able to attain through Wall Street financing deals. For example, LDK’s US issued convertible bonds bear 4.75% annual interest, the same rate as US based Sunpower (NASDAQ: SPWRA) for its convertible bonds. U.S. based First Solar (NASDAQ: FSLR) enjoyed even lower average interest rates through its credit facilities ranging from 4.12% to 4.54% according to the company's most recent annual report.

In its Q4 2010 earnings conference call, Yingli (NYSE: YGE) gave the most insight on the large credit facilities recently handed out by Chinese banks. According to YGE, these large credit lines are not intended for discretionary capital expansion but rather for project financing. Unlike credit given to companies for working capital purposes, solar project financing is secured by hard assets which generate predictable revenue streams -- electricity. Solar project financing is simply a much safer investment for banks, not to mention above 20% IRR in some cases. This may explain why LDK turned to Wall Street financing instruments instead of tapping its 8.9b CDB credit facility.

This may also explain why LDK acquired a controlling stake in U.S. based Solar Power Inc. (“SPI”) earlier this year. SPI’s purchase follows a similar move made in 2009, when LDK acquired a majority interest in Solar Green Technology S.p.A. (“SGT”), based in Italy. These acquisitions expand LDK’s presence to downstream systems development in two major solar markets. If LDK is unable to access its large CDB credit line for capital expansion or to revolve existing debt, it’s possible CDB instead could help finance future projects initiated by the company’s two newly acquired downstream divisions. An arrangement of this nature could potentially account for a large portion of LDK’s fully integrated module production moving forward.

In the end, these three recent financing efforts raised approximately 377m. While it’s slightly short of the company's total 400m convertible bond debt, the amount raised is above the 300m LDK originally hoped to refinance with its initial exchange offer.

Early this year, LDK also announced a 240m convertible preferred shares issuance to a group of investors including several major Chinese banks. Once converted, these shares grant holders an 18.5% stake in LDK’s polysilicon division. In a parallel action, LDK also repurchased the 15% stake of its 15,000 metric ton polysilicon plant sold in late 2009. The transaction appears to be a transfer of ownership from local level to national level investors. LDK also signaled a potential IPO of its polysilicon division in the same announcement. Sources later revealed LDK plans to raise approximately 1b USD in a Hong Kong listed IPO later this year.

By spinning off its polysilicon division, LDK could significantly restructure its debt. First, new capital would be raised from shares sold in the initial public offering. More importantly, LDK could repackage much of its existing debt to its polysilicon division. Most of LDK’s debt is linked to the 1.8b cost of its polysilicon plant. If the debt related to the cost of LDK’s polysilicon divison is also spun off in an IPO, LDK could potentially remove more than three quarters of its existing debt, which would leave the company with a net cash balance. LDK could then focus more on downstream activities, while still maintain a controlling interest in its polysilicon division.

Why would LDK want to delay an IPO of its polysilicon division if it could dramatically restructure the company's debt? An IPO is a one time event with success linked to the ability of the company to sell it to the public. Currently LDK’s polysilicon division is only halfway ramped towards full capacity levels. As a result of underutilization, production costs are also higher than normalized levels. In the last earnings conference call, LDK noted a 43/kg polysilicon production cost coupled with approximately a 60/kg average selling price. Although per kilogram polysilicon gross profit is expected to increase to around 30/kg in the fourth quarter of 2010 from below 20/kg, the absolute level of gross profit from LDK’s polysilicon division is still far from fully normalized levels.

From LDK’s guidance, 1900 metric tons were produced and sold at an around 30/kg gross profit in Q4 2010. Viewing its polysilicon division as a stand alone unit, this equates to 57m in quarterly gross profits. At full utilization, this division is capable of 4500 metric tons of production at 25% lower production costs. If these metrics are achieved, and assuming the same average selling price for a comparative reference, 180m in quarterly gross profit could be generated at full capacity.

Although it’s unlikely current average selling prices for polysilicon can be maintained towards the end of this year, absolute levels of gross profit would still be more than double recent run rates, even when assuming much lower long term contracted silicon pricing. By delaying its IPO until production metrics reached full utilization, LDK would have a stronger selling point for its polysilicon division. That could increase the chances for an IPO success while increasing the funds LDK could raise from the offering.

LDK’s willingness to pay above market rates for its recent financing activities suggest the company is unwilling to alter its corporate planning (capital expansion) for this year. Many of its operating metrics have not fully normalized, as evident in the polysilicon division example described above. With the company’s gross profit potential still under realized, it’s not hard to understand why LDK is willing to pay more to finance its business until fully normalized metrics can be achieved.

Finally, earlier this year LDK raised its Q4 2010 guidance by a fair margin. Revenue guidance was raised from 710-750m to 870-910m. Gross margin guidance was also raised from 24-26% to 25-27%. Shipment metrics were also raised proportionately. With updated guidance, I am also revising my Q4 2010 estimate for LDK as follows:

Core Wafer: 510mw @ .92/watt = 469m
OEM Wafer: 110mw @ .50/watt = 55m
Module: 155mw @ 1.88/watt = 291m
Module OEM: 20mw @ .50/watt = 10m
Polysilicon: 1000mt @ 75/kg = 75m
Total Revenues: 900m

Cost of Goods:
Core Wafer: 510mw @ .62/watt = 316m
OEM Wafer: 110mw @ .30/watt = 33m
Module: 130mw @ 1.78/watt = 231.4m + 25mw @ 1.21/watt = 30.3m, 262m total
Module OEM: 20mw @ .40/watt = 8m
Polysilicon: 1000mt @ 40/kg = 40m
Total COGS: 659m

Gross Profit: 241m
Gross Margin: 26.8%
Operating Expenses: 35m
Operating Profit: 206m

Net Interest Expenses: 25m
Government Subsidies: 3m
Forex Gain: 3m
Tax: 27m
Minority Interest: 5m
Net Income: 155m
Diluted Share Count: 136m
EPS: 1.14

The estimates above represent core operational earnings with minor non-operational items-- such as foreign exchange translations-- factored in. There are a couple other factors which could dramatically alter U.S. GAAP reported earnings. The first involves a 31m arbitration award LDK was granted during the fourth quarter. If this amount is realized in its fourth quarter earnings, it could add around .23 in EPS. LDK may also recognize earnings from stakes in several solar power plants similar to gains Suntech Power (NYSE: STP) reported in its fourth quarter 2010 earnings. If present, additional earnings could result from the activation, operation, or sale of solar power plant interests.

Disclosure: I am long LDK, YGE.

Additional disclosure: No position in STP, SPWRA, FSLR, ESLR.