Near and Far: Connector and Semiconductor Industries
Never the Twain Shall Meet?
By John MacWilliams, Bishop & Associates Inc.

This article is written from the perspective of the connector industry. Its purpose is to discuss similarities and differences between the connector and the semiconductor industries, and to foster greater understanding of the relationship between these two vital markets.

As the photos above suggest, the connector and semiconductor industries have a lot in common. They can also be light years apart in their design origination and specific applications—from ultra-miniature IC sockets to gargantuan locomotive electrical systems and everything in between. (See archived article on Large and Small Connector Companies). These two industries tend to follow each other on order intake and outflow, with ICs being the leading indicator. (See archived article on Semiconductor Industry.) The semiconductor industry is more than six times the size of the connector industry (based on 2008 figures). 

Connector – Semiconductor Top 10

The semiconductor industry—which includes all types of integrated circuits and the dominate companies within the industry—is the engine that drives electronic circuitry. They make equipment innovation possible. Cost-effective silicon integration enables many familiar human interfaces we use every day, such as USB, WiFi, optical mice, and cell phones. Connector designs are a direct result of requirements driven by semiconductor technology and innovation. A recent emerging technology is USB 3.0, and another recent product to hit the market, packed with ICs and more than a few connector products, is the Apple iPad.

A key point to consider is that many of these innovations start with standards groups or consortia started by one or more industry leaders. After the standard is set, and sometimes before the standard is established, equipment starts to roll off the production lines. But it is not until the application is embedded in silicon that costs come down and connector volumes go up. Connector suppliers know this, so even though the semiconductor industry is not a large customer, they must be dealt with in the standards arena. The connector industry could do a better job in working with the semiconductor companies during that defining time, and not take the semiconductor industry’s dominance for granted.

Despite the above, there are many connector designs driven by other factors somewhat devoid of semicon influence. A large number, arguably 50 percent or more, of all connector designs fall into this category. Semiconductors are in the mix, but they are not a deciding factor. In many cases, these are application-specific designs that also garner an important slice of the connector industry’s revenues and profits.

There are reasons for this idiosyncratic relationship between the two industries. One can be found in the nature of connectors: Connector designs are almost infinite, backed by a huge number of applications, and the ability to call on a very large variety of electro-mechanical materials, component parts, assembly processes (stamping, molding) or machined connector housings, or electro-optic parts, glass fiber connectors, and processes. In other words, the sky’s the limit in terms of different types of connector designs, both metallic and glass.

With ICs, the manufacturing process is not piece-part. It is monolithic, and limited to a small number of materials and processes: Si/CMOS, SiGe, and GaAs, etc. This also results in many different semiconductor devices, because companies and researchers are finding new ways to use these processes, including almost unlimited photolithographic capabilities. Photolithography produces feature sizes much smaller than what the connector industry does. Example: MEMS devices, which have <100 micron-sized features that can be similar to connectors, or produce micro-miniature wafer-probe contactors.

Another difference between the two markets is the cost of entry and reinvestment. In the connector business, it is much lower than in semiconductors. This may account for the greater number of companies (300+) making connectors vs. those fabricating ICs (<100). Again, see Small vs. Large.

Yet another significant difference between the makeup of these two interconnected industries: Connector manufacturing is typically turnkey, i.e. vertically integrated from basic materials through manufacturing, with some variations. But, though the IC process is monolithic, an increasing percentage of ICs are designed by one company and produced by another company, such as Walla, an IC foundry operation. The merchant company invests in design, which in itself is not insignificant, while the foundry invests in equipment to defray the material costs over multiple customers. This is the semiconductor industry’s equivalent to outsourcing in electronic equipment manufacture. In fact, there are two levels of outsourcing in the semiconductor industry: Foundry operations and IC packaging. These differences are caused by, on the connector side, a relatively low entry level, and on the IC side, a huge, often multi-billion-dollar capital investment need, not to mention (in either case), the investment and access to IP (intellectual property).

This difference in investment cost partly accounts for the large number of connector companies out there. The top 10 connector companies equal approximately 50 percent or more of the total market (see table above). But that still leaves $20 billion in connector sales for the rest of the industry. Also, if all of these companies were doing the same thing, the market would be much more concentrated, which means there would be far fewer connector companies.

Well, that process is at work. Each year there are mergers and acquisitions in the connector industry. Globalization has resulted in new entrants as well, because as we said above, there is a relatively low barrier to entry.

There is consolidation also in the semiconductor industry. This is caused primarily by huge capital investment costs. It is getting to the point where less than 10 companies in the industry will be able to finance future semiconductor fabrication costs, which are escalating into the multi-billion-dollar range. Some speculate that less than five semiconductor companies are able to absorb the costs of fabricating the 35-22 nm nodes.

In addition to these significant differences, the connector and semiconductor industries are viewed by many as being “joined at the hip.” Wall Street and major business magazines tend to lump these two industries together. Yet, in only one or two cases are there companies who make both ICs and connectors. The commonality is in business and market relationships.

I guess we will have to face the fact that two of the most outstanding industries ever created will march into the future together—even when they don’t know each other’s names.

John MacWilliams
Senior Consultant and Analyst, Bishop & Associates Inc.
John MacWiIliams has been in the electronics industry for over 40 years. His main areas of experience have included: U.S. competitiveness programs, market research studies, authored articles, field sales and management, product marketing management, strategic marketing, new product planning, venture development, advertising and media relations, direct sales, manufacturers representative, distribution sales management, and international marketing. MacWilliams has worked with AMP, Diceon Electronics, TRW, and IRC in marketing management positions. Prior to joining Bishop & Associates, MacWilliams served as the group director of marketing and new product planning for AMP.

MacWilliams graduated from Lehigh University with degrees in business management and engineering.