Jock -
re: My point was that the creation of a new line is in response to declining margins and a declining growth rate in the PC business.
In that case we can dispense with your statement as not germane to the discussion of supply and demand. It is well known that DELL needs to adjust its business model to maintain growth, ASP and margin in the face of increasing commoditization of the desktop PC business. They have publicly stated that they intend to increase share in high end servers and enterprise storage to address this issue, and also that services will play an increasing role in overall corporate revenue. Or is there some connection between your statement and the supply / demand discussion that I have missed?
Moreover, you have failed to address my point that the rules of supply and demand is what fueled the growth of the PC business.
Let's examine this assertion a little. We can define the PC business as the IBM compatible PC, and set the start date as 1981, or we can go further back and look at the dawn of microprocessor-based systems which really were responsible for the term PC.
Looking at the second category first, the Altair 8080-based machine is widely regarded as the first of this class of "personal" machines. The Altair was first announced in 1975. It was closely followed by similar machines from IMSAI and others, with the center of gravity being designs based on the 8080 or Z80 chip. Motorola and others pushed 6800 based designs. Rockwell brought out a system based on the 6500. Base price for all of these systems ranged from $500 to about $1500.
The lack of software kept acceptance of these systems to a minimal level. Microsoft developed Altair Basic which booted directly on a "naked machine" and provided basic OS functions as well as a programming language. This BASIC was quickly ported to a variety of other machines.
By 1979 the industry split into two camps - the "ready to run" systems like the Commodore PET and the Apple, and general purpose systems running programs and compilers over an operating system, the lead one being CP/M. Anyone could use an Apple or PET, but the CP/M systems were more like minicomputers of the day, mostly for geeks. Prices for these systems rose steadily from the $500 machines of 1975. By 1980 both classes of machines were about $3000. Total sales by 1980 was about a million units. So at least in the early days, increasing price went hand in hand with increasing demand. Price was increasing about 45% per year. Demand was increasing at about 200% per year.
This is exactly the opposite of your assertion of the way supply and demand work in the business.
This set the stage for the introduction of the IBM PC in 1981. IBM set their initial price point at nearly twice the going rate for equivalent systems, claiming that 16 bit technology and the IBM name justified a premium. Their initial estimate was that they could achieve annual sales between 100,000 and 250,000 units per year at this price point. In fact, they sold over 500,000 units the first year, and by the end of 1983 were at a run rate of better than 1.500,000 units. The PC industry as a whole was selling over 2,000,000 units, the majority of them IBM compatible machines.
Prices continued to inch up. By the introduction of the IBM AT, ASPs were approaching $8,0000 and by 1986, with Compaq's introduction of the 386, prices for mid-range machines were well over $10,000. None the less, the overall market continued to grow at better than 100% per year.
Moore's law, and the economies of scale driven by the huge volume of components required by the PC industry, caused component costs to begin falling throughout the late 80's and early 90's. By 1992, ASPs for mid-level machines had fallen to less than $5,000. Margins fell slightly during this period also, but remained fairly constant in the low 30s. Despite the decline in average system costs, the rate of growth in the overall market also dropped substantially, down to about 50% growth by 1993.
Here again, the behavior of the PC market for this key 10 year period is opposite to the proposition you present. Falling component costs and related lower costs to end users did not stimulate demand, instead demand growth continued to decrease.
But lest you accuse me of being Rip Van Winkle again, we'd better bring the analysis up to the present day. Decreasing component costs and some margin erosion have continued to bring down ASPs which currently stand at about $1600. None the less, growth in the market has continued to decline, and estimates for 1999 are for growth of about 23%.
So an analysis of the PC industry from its beginning until the present, considering the variables you propose (price and demand) shows not a single instance which supports your contention, and in fact all the evidence shows that if anything, decreases in price have accompanied decreases in growth of demand. It is based on this data that I would suggest that the relationship you propose is simply not important in an analysis of the PC business. |
