Abaci's Common Sense Investing

Of course our Financial Secretary IQT (such named in a column of HKEJ) isn't moonlighting in a PCB factory. Rather it's his family's company Meadville, a PCB manufacturer, which is going through an IPO now. Upon initial read of the prospectus, it seems to be an interesting prospect.

More layers = higher technology = higher margin. This is the rule of thumb for PCBs I learnt from my past clients. But even multi layer PCBs are getting common in China and so we have HDI, high density interconnect, which squeezes in more 'electronic path' than a standard PCB. With denser and multi-layer PCBs, electronics can be made increasingly compact, as we have witnessed the rapid shrinking in weight and size of computers, phones, digital cameras and other electronics over the last few years. The next step is IC substrates which is even smaller in a micron state. As the trend of miniaturisation in consumer electronics isn't changing soon, demand for these high end PCBs is predicted to be much stronger than the overall market.

Manufacturing PCBs is a bloody competitive business and one way to come up ahead is to be the forerunners. In China where there's little proprietary technology this means being the first to adopt foreign technologies in local production. But for this to happen one needs to earn enough reputation and trust among customers before they are willing to transfer their technologies and outsource production. This appears to be the growth story of Meadville who has successfully collaborated with Hitachi and TNCsi, two Japanese firms with more advanced production technologies, and managed to stay in the top end of the market and avoided excessive price competition.

Looking at the breakdown of the latest turnover, about 2/3 was derived from higher-end PCBs (>8 layers), HDI, and IC subtrates. Conventional lower-end PCBs (<8 layers) constituted about 20% and has been on a decreasing trend (down from 48% in 2003). Meadville serves customers in communications equipment (Huawei, ZTE), computer and peripherals (ATI), consumer electronics (Pioneer), mobile phones, automotive, medical and industrial equipment. It's comforting to know many of its customers require 2 years before fully admitting any new supplier. Local/export ratio is about 70/30. Pricing is said to follow a cost-plus model meaning any rise in material cost can be passed on to customers - I'll wait and see.

Capital expenditure is gonna be substantial at $1.2b, $1b, and $650m over the next 3 years. Comparing to the pre-IPO net assets of $1.5b, we're talking about a super expansion phase here! Meadville is gonna be ~3 times as big as it is now. I certainly hope the management is right in their forecast. The IPO proceeds are only enough to cover the first year so I'd expect a lot more gearing or even fund raising in the future. And initial production expenses and depreciation is gonna hurt profit too. Debt is already running high at $1.1b and $700m was paid out as part of pre-IPO reorganization. Financial strength is one key concern, although I don't quite believe the Tangs will risk ruining its reputation with Meadville. (K.S. Lee on the other hand does this trick more and more often now)

I also don't quite grasp the accounting treatment on the giving out of shares by the controlling shareholder to the management and staff. This is supposed to be a transfer of existing shares but it'll nevertheless be expensed off by the company in the 2007 accounts. And the amount will be substantial at $250m.

Profit, as usual in any prospectus, is rising rapidly and Meadville is estimated to have earned $310m in 2006. Capitalization based on the top end price of $2.4 will be $4.8b or $5.0b with over allotment, giving a 2006 p/e of 15.5-16.1x. This is not cheap at all for an industrial and certainly the Tang family name is sold at a premium. P/B at IPO will be around 1.8x. Valuation is built upon success in future expansion and management quality. I'm gonna have a dip in this IPO (I won't receive a lot of shares anyway) and see my luck, just for excitement in this boring market when only China Mobile and a few are rising. Meadville is certainly worth another look at should its price becomes more affordable down the road.

Similar companies listed on market are Daisho Microline and Hannstar but they are not as diversified in customer base. DM is focused on mobile phones and Hannstar is on notebook computers. But they are selling at much cheaper valuation at maybe 6x and 10x respectively. Or WKK the PCB machinery trader is even cheaper. These may be more interesting bets.

DISCLOSURE: I have applied for 3313 and hold 532 at time of writing. I don't hold 567 or 667 at time of writing.