On June 28th, research firm Muddy Waters (MW) released a critical report on Spreadtrum Communications (SPRD) and disclosed it was short the stock. The stock had violent trading the rest of the day; it traded from $13.20 down to $8.60, then back up and closed down slightly (click to enlarge images):
This price action had me wondering who was correct: MW (bears) or Jefferies (bulls)? I had followed an earlier piece of MW research and found it credible, so this led me to look closer into SPRD's financial statements. I went to the SEC's website and pulled up SPRD's last annual report.
I have found some preliminary problems with different figures being reported on SPRD's financial statements. On page 7, where it discloses selected financial data, it lists "cash and cash equivalents" as being $82,195 (in thousands) as seen here, highlighted in red:
Then, on page F-4, it reports there being $81,185 (in thousands) on its balance sheet and statement of cash flows, as highlighted in red again:
This is a difference of just over $1 million! One can also see that the amounts for 2009 match but 2010 does not.
Now maybe SPRD just made a mistake. However, it would have mistyped both a 1 into a 2 and also an 8 into a 9.
I continued my inquiry by looking at how its cash flows match up to changes in the balance sheet.
Cash and cash equivalents
From 2010 to 2009, the company increased its cash and cash equivalents + restricted cash by $52,376,917. If you add together its cash and restricted cash, from the statement of cash flows, the amount totals $51,367,515.
The company showed just under $8 million going into the restricted cash account on page F-7. However, when one looks again at the balance sheet (just under cash & equivalents), the restricted cash account was $11,496 million in 2009 and increased to $20,496 in 2010; an increase of $9 million. That is an increase of over $1 million more than the statement of cash flows.
It would appear SPRD is having problems keeping its numbers straight.
Also interesting is the large drop in the cost of revenue sold, in 2010, as show by the common income statement on page 57. A drop of 7.6% from 63.6% to 56% seems very large, and again when compared to 2008 with cost at 61.9%.
The company attributed this to:
Cost of Revenue. Our cost of revenue increased by 189.8% to $193.9 million in 2010 from $66.9 million in 2009. The increase was primarily attributable to the significant increase in our sales volume which led to an increase in our semiconductor production cost, sales rebates, royalties and other expenses, partially offset by a decrease in inventory write-down and a decrease in unit semiconductor production cost as a result of reduced average purchase price in connection with our increased purchase volume. The amount of inventory write-down was $1.9 million, a decline of $3.5 million from $5.4 million in 2009.
I do wonder how sustainable a large decrease in inventory write-downs is when sales are reportedly increasing.
Cost of goods sold (revenue) = beginning inventory + purchases - ending inventory
Since inventory is being held higher, according to MW's report and SA article, this should depress COGS and let SPRD report a higher gross margin in the short term. However, this process will likely revert in the long term. Also, will the company be able to keep operating expenses at only 24% of revenues and still grow the business?
The above evidence makes me have the opinion that MW is likely on to something again. SPRD has to explain the differences on its financial statements. Until then, investors should stay out of the stock; speculators can short and expect a lot of volatility. It looks like SPRD has dropped its signal.