Quality Is the Answer to the Labor 'Crisis' by Paul A. Strassmann Computerworld June 8, 1998 The shortage of systems personnel is a temporary phenomenon. A historical study of all shortages - whether they were for food, gold, oil, land or fish - teaches us that they are aberrations caused mostly by temporarily excessive demand. The economics of rising prices will always generate corrective measures that ultimately deliver a sufficient supply. Shortage and self-interest Vendors are happy about any "shortages" because they make selling easier and prices higher. The recent discovery that there was a shortage of systems and programming personnel was welcomed by just about everyone who stands to benefit from it. For example, the Information Technology Association of America, a trade organization that represents primarily the computer industry, favors more liberal immigration quotas for computer experts. The National Research Council and the Computer Research Association, both of which mostly reflect the views of academia, favor more subsidies for computer departments. The Software Publishers Association, which represents lobbies for Microsoft, Oracle and others, advocates spending public money to boost the supply of programmers. Computer people love the news about scarcity of talent because it enhances their sense of importance and legitimizes requests for larger-than-average raises. For someone aspiring to become a computer professional, enter the head-hunting business or form a software company, it would be useful to understand how real and sustainable the current shortages are. Shortages Don't Last There's a simple way to eliminate shortages: cut waste and improve productivity. There's no lack of examples. Every few years New York runs into alarming water shortfalls because much of its supply goes to waste from leaking pipes and poor supply management. The land-rich Soviet Union experienced chronic shortages of food while most of the crops were left rotting in fertile fields. Only 15 years ago, the price of gold was expected to rise to more than $1,000 an ounce on account of shortages. Speculators lost money when a drop in demand brought the price down to less than a quarter of what was expected. The yesteryear scarcities of oil were supposed to lead to prices of more than $100 a barrel. Nowadays, oil costs less, in real dollars, than when the scarcity scare started. The food famines and fuel shortages that were accepted as a reliable prediction in the 1960s have now been replaced by surpluses in all raw materials, with prices dropping relative to the rising global personal income. There are many reasons to believe that the IT labor "crisis" will follow this pattern: resistance to rising labor costs, automation of programming tasks, standardization of applications, adoption of total quality methods, institution of systems engineering disciplines and certification requirements. Management will resist hiring more systems professionals as the solution to their problems. Instead, they will start boosting ways to deliver greater productivity through improved software quality. The Demand-Management Proponents Early this year, a task force report focused on how software productivity and quality relates to the U.S. information technology workforce shortage. Howard Rubin, chairman of the computer science department at Hunter College in New York, chaired and authored that effort, which included contributions from other software quality luminaries such as Barry Boehm, Jerry Grochow and Capers Jones. The report concluded that U.S. software productivity is unsatisfactory and that it has been declining since 1993. The management of software continues to be unacceptable, with only 6% of organizations meeting the criteria of having "defined" software management processes in place. The report's authors concluded that increasing software quality, as well as reducing the "destructive feedback loop" from excessive employee turnover, would relieve the current tendency to compensate for unsatisfactory performance by adding more people. Contributors to the software quality and productivity report favor increasing the supply of skilled systems personnel. But they also identify the following practices that are necessary to achieve greater efficiencies: software reuse, systems integration standards that aren't vendor-specific, simulation methods for capture of customer requirements, and formal software testing processes and certification of software for conformity with generally accepted standards. Of those measures, the reuse of high-quality software components would yield, by far, the highest cost-reduction potential. Implications for Management Rising prices, short technology life and chronic systems failures are rapidly reducing executive tolerance for waste. Boards of directors are recognizing that the current shortages of qualified systems personnel are largely self-inflicted, with the year 2000 fiasco offering the most immediate proof of software mismanagement. The prevailing waste in managing software as a corporate asset is scandalous. According to a study by Jones, chairman of Software Productivity Research, Inc. in Burlington, Mass., canceled projects consume more than 15% of all software efforts. Furthermore, about 60% of the U.S. software workforce is engaged in fixing errors that are largely avoidable if total quality management practices were applied. If those numbers are correct, proper software management practices could lead to more than a 1 million personnel surplus. Therefore, I don't believe that we suffer from a 350,000-person shortage, as has been alleged by those who are looking for more billable bodies to cure a situation that has been caused by mismanagement of valuable people. The present shortages of systems personnel in the U.S. will vanish as normal growth, global competition and improved systems management practices alter the current conditions. That will happen within the next decade, for sure. One should not count on the persistence of premium prices and the imbalance between demand and supply to make decisions that bank on that flawed assumption. Strassmann (ceo@stacorp.com) says certification of reusable software components to high-quality standards will displace large amounts of handcrafted programming.

Go back up to the Strassmann, Inc. home page.