A case study on transition from macro- to micro-level,
or When the tech stocks will rebound in earnest.

Y. B. Karasik,
Thought Guiding Systems Corp.,
Ottawa, Canada.
e-mail: karasik@sympatico.ca

The widely advertised explanation of the current economical crisis is as follows: businesses stopped spending. But why did they stop it in the first place ? There is no clear answer to this question in the media. Partial explanation refers to the burst of .com bubble. It is invoked every time when one tries to explain the crisis in the telecom sector.

Allegedly, after the demise of many .com companies, the telecom service providers were left with excessive capacity. As a result, they stopped capital spending and thereby triggered troubles of telecom equipment makers (Lucent, Nortel, Alcatel, Cisco, Ericsson, etc) Their stocks crushed. The waves went on and on: to semiconductors (because the less of them are needed when the less telecom cards are needed); to semiconductor equipment makers, etc. etc.

Immediately a macro-economic remedy was proposed for the crisis: to lower interest rates and thereby stimulate companies to buy new equipment. As is known, it did not work so far. Nobody buys if he/she does not need (regardless of price).

Meanwhile, the above macro-economic analysis of the causes of the situation is not corroborated by ... No, not by micro-ecomonical analysis, but by really micro-level analysis of the problem. What is this "real micro-level analysis" ? It is an analysis from the standpoint of a level, which is really "micro" relative to the level of economy. What is the "micro" for economy ? The underlying level of technology, of course.

Thus, let us take a look at the root causes of the crisis as they are seen from the standpoint of technology.

For almost all XX century, the telecom networks were dominated by voice (or phone) switches. They evolved gradually over decades and reached near perfection by the late 1980s when the digital voice switches were eventually perfected after almost two decades of improvements since the 1970s. These switches were designed to hold connections for almost 1 hour (who speaks over the phone longer ?); to connect 100,000 subscribers (who needs more ?), etc. They had reduandant (i.e. stand-by) cards, and many other elements of reliability. And when this paradise was reached, a small nuisance appeared ! It was internet.

People started to connect computers to telephone lines and browse internet for hours, not minutes. The switches were not designed for this. They could not hold connections for hours. They dropped them and users got angry and demanded compensation. Moreover, even if switches managed to hold connections for hours, they could not transmit files over the lines fast enough. Files, especially video, requires much more bandwidth than voice.

In short, the telecom service providers started to lose money because of the unsuitablity of the equipment and the anger of the subscribers. The providers started to bang on the doors of the telecom equipment makers (Lucent, Nortel, etc.) asking for help.

The latter ones started to hastily design ATM switches (suitable both for voice, data, and video transmission), IP swicthes, optical switches, optical networks and cards, etc. etc. Not that this work was not conducted previously. But in the 1980s, it was mainly research on a small scale (without anticiapting demand from the market any time soon). In the second half of the 1990s, it became a feverish design, though. The telecom equipment makers came up with "answers" to the providers demands !

The latter believed them and started to buy the new (and expensive) equipment in huge quantities. Very soon they discovered that the swicthes were extremely unrelialble, behaved unpredictably, etc. They used to lose all connections of a card very often. The stand-by cards could not recover them as well. Switches needed to be reset often, etc etc. All this created such a headache for technicians and maintenace teams in general, that some providers started to demand removing the new equipment and compensation from the manufacturers. Providers were losing money on them rather than making. Firstly, because they had to pay compesation to unsatisfied customers. And secondly, because they had to hire an army of technician to support the equipment.

Thus, telecom providers have really overspent. But overspent on junk ! There is really over-capacity. But overcapacity of junk. Because no good and revolutionary telecom networks can be designed and perfected in a few years. For voice networks it took decades. Why should it take less for ATM swicthes, optical internet, etc. etc. ?

Here is the root cause of the telecom crisis, and not the burst of the .com bubble. No interest rates (whatever low they can go) will solve the problem. Only time will. Thus, do not expect rebound in the tech stocks in earnest any time soon !