Thursday, November 13, 2008
Electronics and the economy: a view from Electronica
The giant Electronica trade show gives an interesting window into the global electronics industry. This year, that view could be critical. But it is ambiguous.
Perhaps unlike much of North America, there are no manifest feelings of impending doom here. Business is brisk, the halls are for the most part full, both of exhibitors and of attendees, and the talk is about new products, not layoffs. So how does one explain the contrast to the drumbeat of layoffs, slashed guidance, and fear that has been grabbing headlines in the US?
The simple, but insufficient answer, is that one doesn't go to a trade fair to frighten one's customers. Certainly true, but I think that misses some important distinctions. These differences involve the market focus of the companies here, and the way the typical Electronica exhibitor—as opposed to, say, Intel or Qualcomm—engages the market.
To explain, think of the electronics supply chain for a moment as a giant delay line, stretching from end-users at one end, through OEMs, component manufacturers, and eventually back to basic technology creators at the other end. This basically is the path money follows in moving from my credit-card account to a research lab.
As a zero-level approximation, the financial crisis has had two impacts on this delay line. It has imposed a falling step function on the input of money from the consumers into the economy--by making off with their investments, undermining their houses, threatening their jobs, and generally scaring the wits out of them. There has also been a secondary, nearly simultaneous impact on just about every other stage of the pipeline: the freezing of the credit markets has made it more difficult for companies to borrow for operations or new investments. Fortunately, for most companies in the electronics supply chain, this impact appears to have been smaller than the impact of consumer spending, for a number of reasons.
This model uses a delay line because there is a finite time delay between the impact on one stage and the impact on the subsequent stage. So, for example, when consumers headed for the hills, OEMs in consumer electronics felt the impact almost at once. And indeed, there have been stories circulating here about substantial layoffs of factory workers in China. After some delay, these impacts are now starting to ripple into the manufacturers who supply components directly and exclusively to the consumer-market OEMs. After a further delay the impact will arrive at the door of the component suppliers' suppliers, and so on.
That's where one difference originates, I think. Most of the companies exhibiting at Electronica are from the middle of the supply chain. They are suppliers of modules—wireless, power supply, and so on—or IC suppliers, or even more fundamental suppliers of components and materials. After all, relatively few consumer products are built in the EU, so there's not much reason for the people at the very end of the supply chain to be here. The customers of the organizations that are at Electronica are more likely to be industrial companies than to be consumer electronics giants. Hence, they will see the leading edge of the recession later than the headline consumer electronics companies and their key suppliers.
The second point here is that because many of these companies are further up the supply chain, their products are less application-specific. These vendors are often targeting several different end-markets with the same technology, so that when the ripple hits one set of customers who are close to consumers, there may not yet be any impact on customers in later stages in the chain.
Clearly this model raises a key question. Is the connection between stages in the delay line a pure delay element, an attenuator, or an amplifier? If it is a pure delay element, and economic measures can induce consumers to come back to the market in modest numbers, many companies may see no more than one or two soft market areas at a time, as a relatively narrow pulse ripples through the supply chain. If the consumer boycott is longer, the impact will be progressively more severe for diversified suppliers.
If there is attenuation in the coupling—for instance, if each stage in the line has enough market diversity to moderate the impact they pass on to the next stage—then even if the falling edge is severe at the beginning of the line, the ripple could be relatively minor by the time it reaches, for instance, active and passive component suppliers.
If there is amplification—as happened, for instance, in the dot-com bust because inventory uncertainties caused each successive stage to panic—the end result could be a bistable system that has to be kick-started back into operation after the drop in consumer spending has clamped off each successive stage. Or the whole chain might require restructuring.
Perhaps I'm being optimistic, but it seems to me from walking the floor at Electronica that the best model may in fact be mild attenuation. Companies have for the most part been conservative with leverage and are not running out of operating capital, although they are certainly spending conservatively. I heard no reports of designs being abandoned, for instance. Companies at this show—even the small ones—are often quite diversified. And today, fortunately, the level of inventory visibility and control seems to actually be what everyone said—incorrectly—that it was in 2001: excellent.
This model would explain why business is going on, conservatively but actively, here in Electronica. And it suggests a level of cautious optimism is in order for the global industry as a whole, with some definite danger spots in consumer electronics and capital-intensive, high-ticket areas such as automotive or fab equipment that do depend on access to credit markets. But it is a model, and like any other model is not predictive, only suggestive, without calibration. Here's hoping we don't get enough recession data to finish calibrating it.
edn.com
0|0
P.S. This one's for you... Mr. Niceguy767. :) |
