September - October 1999
eMail the Editor
How the Economics of Business Cycles Impact the Electronics Industry- By Harvey Miller
Short-term business cycle downturns are moderated by long-term uptrends and deepened by long-term downtrends. In other words, while a good economic environment doesn't guarantee success for all individuals, companies or nations, it helps.
The speed of the current recovery in the world semiconductor industry is a testimonial to the influence of the powerful "long wave" uptrend which began around 1992 or 1993 and is "scheduled" to continue until about 2017 (in the U.S.-more about other regions below).
The fast bounce-back of such nations as Korea and Japan is another testimonial to the strength of a positive economic environment. It's good to know that the direction of the long-term trend is up-but for how long?
That is a more difficult question.
Economists N. Kondratiev and J. Schumpeter brought powerful insights to help answer the longevity question.
Kondratiev created his important work in the early 1920s. He studied cycles from the late eighteenth century as capitalism became the dominant world system. The implication of his work was that capitalism has more than nine lives-not consistent with a dogma that capitalism was about to be buried. Ultimately, Kondratiev disappeared into one of Stalin's gulags.
Schumpeter, theorizing in Austria during the early 1930s, was surrounded by the rise of tyranny to the right and left. The clarity of vision he brought to his studies of the dynamics of the cycles of capitalism was a valuable legacy from those troubled times.
We shall apply their ideas to our time, specifically the long waves that followed 1948 and on to the twenty-first century.
Below is our simplified chart of the major trends of the twenth century. The Dow Jones Industrial Average in constant 1948 dollars is shown at each half-cycle point. (No DJIA is shown for 1903, because the Bureau of Labor Statistics began to compute the Consumer Price Index only in 1913.)
The price gauge which had always marked uptrends reversed after the early 1970s. The 1970s and 1980s were a period of downtrend, but instead of deflation (which had always accompanied previous long wave downtrends) there was extreme price inflation.
Why did the historical pattern shift? The current uptrend, which began in the early 1990s, is also perverse, based on past norms. Deflation is the hallmark of the current long wave uptrend. When the deflationary forces give way to price inflation, we will know that the long wave uptrend is coming to an end and that the long wave downtrend is due.
But what reversed the old pattern in the 1970s and after? The answers illustrate the dynamics of long waves, both their common and unique elements.The "Cold War" Era
On March 5, 1946, Winston Churchill rang in the Cold War era with his Iron Curtain speech at Fulton College in Missouri. The human and material rubble of war still had to be resettled and cleared, and the long wave of 1948 to 1993 was about to begin (J. A. Schumpeter, Capitalism, Socialism, and Democracy, 1942).
Here are summary features of its up- and downtrend phases:
In that 20-year long wave downtrend, 1972 to 1992-93, new industries emerged: ICs, microprocessors, DSPs, memories, HDDs; then computer mainframes and minis, leading to the PC; the beginnings of networking, the protocols and initial hardware of the Web. This is "creative destruction" in action. Major old industries mature; some die. Capital flows to new ones.
The key element that launched the new long wave cycle was victory in the Cold War, highlighted by the fall of the Berlin wall on Nov. 9, 1989.
All creative forces and energy were free to focus on peace. A major implication for the long wave uptrend was that only one superpower was left standing. The Third World lost its commodity price bargaining leverage. U.S. prosperity and Third World recession are opposite sides of the same coin. But economic nature abhors superpower vacuums, and Europe soon became the first to fill it. China will build the necessary infrastructure but won't attain full superpower status in the current half-cycle uptrend.Contract Manufacturing
The hallmarks of the period are well known: price deflation, innovation, short product life. These factors have induced OEMs to concentrate on design and to farm out low-margin operations. Similarly, a large group of fabless semiconductor manufacturers has emerged whose reason for being is design.
Contract manufacturing in many forms is a product of the 1990s. The impact is different on three major segments of contract manufacturing for reasons specific to each.
We'll concentrate on semiconductor packaging and test.
Since the early days of manual wire bonders, the Pacific Rim has been the production focus. Until the 1990's, packaging was primarily captive, an extension of wafer fab, also almost entirely captive. But new, smaller, denser array packages have emerged to meet equipment needs for smaller footprints, IC needs for higher lead counts and assembly (package and board) needs for higher yields.Onshore Opportunities
Automation plus fast response seemed to open onshore opportunities. Sound in theory, but theory ignored the fact that the capital invested in automation requires steady return, as does relatively high priced U.S. staffing. The semiconductor industry is not very accommodating in offering either steadiness or predictable product life.
There was merit in the idea of onshore IC packaging, but the irony is that the U.S. operations are now merging with offshore ones to offer the best of both worlds. For example, there are liaisons between Abpac and Siliconware (Taiwan), IPAC and OSE (Taiwan), and ISE and ASE (Taiwan).
Two new factors will influence the future of semiconductor packaging: (1) As chips become inexorably cheaper, flip-chip on board will become more attractive for miniaturization and cost reduction. In portable equipment especially, this trend will eliminate packages. 2) Wafer-level packaging is a reality. Ultimately this will put the packaging operation at the end of the wafer fab line-and it will all become captive again or go to the wafer foundries.Harvey Miller secured his bachelor's in economics from the University of Michigan but was sidetracked before he could receive his master's from the University of Chicago. He received an "A" from economist Milton Friedman-which, Mr. Miller says, will now probably be revoked. Contact him at firstname.lastname@example.org.
Chip Scale Review o 7291 Coronado Drive, Suite 8 o San Jose, CA 95129 o Email: email@example.com
|© 1999 ChipScale REVIEW|