Deswell Industries (DSWL, Financial) is a Chinese company that manufactures plastic parts, printed circuit boards and other components. Last year most Chinese companies that came public through a reverse takeover proved to be frauds, making it clear that there are big risks involved in investing in foreign countries where it is not easy to check if a company is telling the truth, and there is no system in place to prosecute fraudsters. So there are good reasons why people are staying away.

I have looked at several Chinese companies in the past year, initially trying to figure out if there could be value. Those RTO-companies became very cheap on every imaginable financial metric after the fraud allegations came out. Initially, being way too naive, you see a lot of arguments for the short case that make sense, but also a lot of arguments for the long case (you would for example expect that big hedge funds do actually do some due diligence when they decide to invest), and you figure if it would be 50/50 fraud/real it would still be an insane opportunity. With a bit of luck I managed to avoid disaster because the companies SEC filings (In this case talking about CCME) made no sense at all if you dug in some of the agreements the company made.

But if I’am honest: I only managed to avoid CCME because of the stupidity of the management, and not a lot later Sino-Forest showed just how elaborate, long running and large scale fraud could be. So with that in mind: how could you ever become comfortable with any Chinese company?

When is fraud not possible?

To answer the fraud question you have to ask: what is the goal of fraud? That’s easy, it’s stealing money, and money has to come from somewhere and someone. For a company there are basically three ways to attract money and defraud investors:

• IPO the company: making it possible for insiders to cash out from a non-existing or materially misstated business.
• Issue new shares: it’s not a lot different from the first option, but the great thing is you can just keep issuing new shares if you can find buyers.
• Issue debt: see above

By definition every publicly traded company has to come public at some point in time, so there is always a mechanism available that makes fraud a possibility. If a company went public a long time ago, and didn’t issue any debt or equity since it would be unlikely to be a fraud, but you can’t be sure. Maybe they are just too busy spending money, and want to leave the option open to get a refill if needed.

So what you also want to see is that the company is returning cash to share holders. Just returning a little bit of money is not enough. Paying a few million in dividends, or having some insider buys, and then trying to raise hundreds of millions in debt or equity would be a wonderful model that Bernie Madoff would approve. But if a company would raise for example a total of 10 million dollar, and then return 20 million dollar in dividends, you can be sure. It would be the worst fraud ever.

So the idea is really simple, and my excuses for wasting so many words on it, without even having started to talk about DSWL, but I have good reasons. The idea was posted on Whopper Investments a few days ago, and you see that no-one is willing to buy the company, including the author of the article, because it’s supposedly too hard to evaluate if the company is a fraud. And that’s exactly why it could be an opportunity: no-one is willing to buy some unknown Chinese company. I guess there is a thin line between being contrarian and not being open to other people’s opinions, but when everbody is staying away it might be a good time to go against the crowd.

How does Deswell do on these criteria?

Based on the previous discussion you can probably guess where this is headed, so here are some of the facts:

• The company came public in 1995 through an IPO (not through a RTO)
• The company issued a mix of shares and warrants, raising approximately 20.5 million dollars between 1995 and 1998. No equity has been raised since (if you look at share count numbers from old filings note that it has executed a three-for-two stock split in 2002 and also 2005).
• The company started paying a dividend in 1996, and has been paying out money every year since, for a total sum of approximately 90 million dollar.

Of course not all that money has been returned to outside investors since insiders own part of the outstanding shares as well (and still do), and it’s a bit harder to track how much insiders have exactly made by selling shares on the open market. The filing history only goes back to the year 2000, and while the company does offer some information on 1998 and 1999 it’s not enough to figure it out exactly. Insiders owned 64% of the company in 1999, and sold a lot of shares in 2007, keeping an interest of ~20% since. With the stock around 10$ that year this would probably have resulted in a gain of ~60M dollar.

So if we add this up it’s hard to believe in the fraud case: the company is simply returning way to much money to share holders compared with how much is coming in. You don’t have to have special insight or skills to evaluate this case, a little bit of common sense is in my opinion really more than enough!

How cheap is it?

Cheap. The cheaper something is, the less time I spend on trying to value the business as accurate as possible, figuring that can wait until the stock price rises significantly. But just to get a ballpark figure: the company has a 35M market cap while holding 36.7M in cash and equivalents. And then of course we do have more assets such as accounts receivable, inventories and PP&E, and there is a business that is actually generating positive cash flows as well (total book value is 122M). Just to give a rough idea some historical key statistics from Morningstar (does not include the results of the latest quarter that were, at the time of writing, just released):



As is visible the business is in a decline: margins have been dropping because inflation in China is high and labour costs are rising (in the last few years the minimum wage was raised with 20%, twice!). DSWL produces commodity products: it does not have a real competitive advantage. The company used to be way more profitable, but just being in China is not as big of an advantage anymore, and maybe the future for commodity producers are even cheaper countries such as Thailand or the Philippines.

But I don’t think you should be overly pessimistic on China, Apple for example believes that it wouldn’t even be possible to move their production back to the USA, not because labour costs are higher in the USA, but the required workforce is not there. A quote from a NY Times article titled “How the U.S. Lost Out on iPhone Work”:

Apple had redesigned the iPhone’s screen at the last minute, forcing an assembly line overhaul. New screens began arriving at the plant near midnight.

A foreman immediately roused 8,000 workers inside the company’s dormitories, according to the executive. Each employee was given a biscuit and a cup of tea, guided to a workstation and within half an hour started a 12-hour shift fitting glass screens into beveled frames. Within 96 hours, the plant was producing over 10,000 iPhones a day.

“The speed and flexibility is breathtaking,” the executive said. “There’s no American plant that can match that.”

Assuming that the company is worth book value ($6.6/share) is probably a bit optimistic, but lets say the company is able to do 80M in revenue with a 15% gross margin. Deduct ~4M in maintenance capex, take a 30% tax rate and multiple that with a 8.5 no-growth ratio and you will get a business value around 50M which would imply a share price of $5.3 after added back the cash balance. It’s rough, but with the shares trading at $2.14 you don’t have to be overly precise.

Conclusion

The thesis for Deswell Industries is beautifully simple, but you have to be comfortable with the logic behind it, because buying a company that looks cheap but isn’t real is probably not going to end very well.

While finishing this write-up the company gave another hint that it is undervalued and not a fraud. DSWL announced today that it has authorized a share repurchase program for $4M over the next two years. The program could create significant shareholder value and it also shows the willingness of Deswell to return money to shareholders.

On a scale from 1 to 10 I give the stock a 8.5: I’m confident that the company is not a fraud, and that it’s cheap. It’s not a great business.

A final note about portfolio diversification: I also have a position in 0684.HK (Allan International Holdings), and this company also is very exposed to the risk of rising labour costs in China. While I’m not particularly worried about this risk, I certainly don’t want to have a lot of exposure, so I have decided to reduce my stake in Allan International (have to figure out how to track this nicely on the portfolio page).

Disclosure

Author is long DSWL, 0684.HK