Consultant usage by large corporations and governments has exploded since the 1970s. In both types of organization, tough decisions are outsourced to consultants so their victims can blame the consultants rather than management or bureaucrats. Consultants are also used to implement ever more complex IT systems, their natural incentive being to stretch implementation projects for decades and make the systems as unusable as possible. However, AI is coming, with its ability to accomplish drone-like tasks previously carried out by consultants, and its ability to take the blame when difficult decisions must be made. This mass extinction of function couldn’t happen to a nicer profession!
Management consultancy has overall a fairly unhappy history, albeit with some exceptions. The first management consultants, Arthur D. Little, founded in 1886, specialized in solving companies’ scientific or technical problems. That was a sensible approach; companies as they grew in an industrial economy moved into new fields of activity in which their top managements lacked technical expertise. This form of consultancy did not require a behemoth client; even quite small companies could use its services. Arthur D. Little survived for more than a century (I interviewed with them – unsuccessfully — in 1978) before filing for Chapter 11 in 2002. (Nominally, it still exists, having been bought out by some of its partners.) Among its successes were setting up General Motors’ R&D lab in 1911 and designing the NASDAQ trading systems for London and Tokyo.
The first consultancies to focus on their clients’ business processes were the engineering consultancies of Frederick W. Taylor and Philip and Lilian Gilbreth, which examined their clients’ production processes in minute detail and prescribed more efficient ways of operating – thereby making production line workers’ lives thoroughly miserable. Their methods were not universally appropriate. One company that rejected them was Baldwin Locomotive Works, a partner of which pointed out to the U.S. Commission on Industrial Relations in 1914 that Taylorism added an extra layer of bureaucracy and worked well only in mass production companies with mostly unskilled labor forces. For Baldwin, where each locomotive manufactured was individual, technical innovations were frequent, products were highly diverse and the market cyclical, the Taylor system was impossibly rigid.
Even if Taylorism was useless to it, Baldwin is a sad case of a company that could have used Arthur D. Little. It employed no MBA professional managers, promoting managing partners from its highly skilled and intelligent workforce. Its top management were thus true experts in steam locomotives, but lacked industrial breadth, and by 1910, Baldwin’s market was beginning to change. The 1906 Hepburn Act, fixing railroad prices, had severely limited railroads’ profitability and consequently their innovation and capital investment in new locomotives. Meanwhile electric locomotives (the first electrified main line in Italy was developed by Ganz Works’ Kálmán Kandó in 1902) and later diesel locomotives (appearing gradually in the U.S. from 1925) would replace Baldwin’s steam locomotive technology. Arthur D. Little, around 1910 or even in the early 1920s, could have told Baldwin of the new competing technologies and helped it to adapt to them; as it was the company faded away, ceasing to manufacture locomotives in 1956 and closing in 1972.
The growth era in management consulting began in the 1960s, when specialists such as McKinsey and Boston Consulting Group (BCG) began advising companies on their strategy, developing various cute diagrams and matrices so that even the slower corporate managers could understand their recommendations. To provide this service, the consultants sucked in the top graduates of the Ivy Leagues and the top business schools, who were painted a flattering picture of their future activities in restructuring the Fortune 500 single-handed, without mentioning the tsunami of utterly tedious presentation and report preparation that would actually be their near-term fate for 100 hours a week. The consultants then marketed their staffers as the best and brightest, implying as subtly as possible that they could run intellectual rings round the dozy corporate managements they advised.
The consultants’ efforts were greatly aided by the era of rapid inflation, low profitability and declining share prices (in real terms) of the 1970s. Shareholders enraged by their continual economic impoverishment could be assuaged by the assurance that management was being advised by McKinsey or BCG and would shortly produce a magnificent new strategy that would revolutionize the company’s position. The strategy consultants knew nothing about the businesses concerned, but their superb qualifications, it was thought, made them capable of managing any business after only modest exposure – a theory that was exploded when the 1960s conglomerates, managed in this way, all went bust. Through their strategy consultants’ efforts, 1970s managements avoided the swift and unrewarding departures they very often merited.
Consultants’ services were then greatly assisted by the movement in the 1980s and 1990s towards a frenzied corporate acquisitions and disposals market, boosted by the “value-based management” fad, which normally recommended a gigantic corporate restructuring and mass firings of ordinary workers, while management rewarded itself with Pharaonic grants of stock options. Naturally, the consultants were invaluable to managements, as they could be blamed for the redundancies – if the giant brains at McKinsey or BCG thought them necessary, who were ordinary employees to differ?
The incentives around consulting were already skewed, given the consultants’ need to secure ever-greater assignment loads to support their metastasizing staff, but they became truly damaging with two new businesses, dating roughly from the 1990s: public sector consulting and the construction of IT business process management systems.
Consultants seeking to maximize their income from the public sector have two major advantages: it is generally impossible to measure accurately the value of their work and the people purchasing it are spending, not just money for which they have only nominal responsibility, as in large corporations, but money from an entirely external source for which their responsibility is in practice zero. Consequently, consultants and their public sector clients conspire to produce metrics showing magnificent effects from the consultants’ work, while in practice ensuring that the flow of fees continues unabated and nothing is achieved that might disturb that flow. If a consultant solves a public sector problem, its fees immediately dry up; therefore, it has no incentive to do so. Unlike with lawyers who have similar mis-incentives, there are generally no judges around to blow the whistle. In Britain, the explosion of public sector consultants was the work of Tony Blair’s government; like most other things that government did, its hugely damaging effects have only become apparent in succeeding decades.
Infotech is an especially difficult case, because only specialists, not untrained management, can discern what is being done. It is thus possible for unscrupulous consultants to amplify the complexities of IT projects, causing their costs to explode. Only 2.5% of companies successfully complete their IT projects, according to a PricewaterhouseCoopers study of 10,640 projects, while one in six large IT projects incurs cost overruns of 200% or more, according to a Harvard Business Review study of 1,471 projects. IT project failures are estimated to have cost the EU 142 billion euros as far back as 2004, indicating as above that the problem is even worse in the public sector.
Artificial Intelligence will not solve all these problems, but it will make a substantial dent in them, thereby reducing the market for management consultancy services. Much of the report and presentation work done by junior consultants in strategy consultancies can be produced by a skilled AI technician in minutes, at the touch of a button. Even more important in the IT sector, AI is now capable of writing software adequately, and its cost and error rate in doing so are a tiny fraction of the cost of human software – for one thing, de-bugging becomes a trivial task. Initially, AI will reduce consultants’ costs and the need for junior consultants, but clients will of course demand that those savings are passed on. The shares of large publicly traded consultancy companies are down 30% or more this year, with the companies focused on IT work especially vulnerable, because AI’s effect on their output is especially great.
The market for consultancy will not disappear, but it will shrink drastically. More and more IT projects will be handled in-house, without the assistance of outside consultants, as one competent IT manager will be able to direct the AI to solve problems that arise. Strategy consultancy will continue as today – Volkswagen will still need to hire McKinsey, so it has an outside “expert” to blame when it fires 100,000 employees. However, the ability to leverage one strategic study into a multitude of implementation projects will lessen, as AI will assist greatly with the details of implementation. Middle Management is an endangered species in the next 20 years, but consultants even more so because they are more expensive and lack detailed knowledge of a particular company’s operations, as AI can be trained to have.
The most likely consultants to survive are those performing the functions of the original Arthur D. Little, using their detailed knowledge of increasingly widespread and complex technological developments to guide companies that need to make strategic technological diversification moves to avoid their own obsolescence. The ideal consultant will then not be the fluent, snappily dressed, superbly groomed BCG “expert” who impresses customers with the brilliance of his mind, but a scientist or technician with deep insights into the direction of technological development. He will probably wear overalls, not a Hermès suit, and carry a wrench as well as a laptop.
Consultancy is a largely parasitic profession, caused by the gross excess in today’s society of highly credentialed college graduates, with no mechanism for sorting them for competence since the infamous Griggs vs. Duke Power 1971 Supreme Court decision effectively banned the use of IQ tests or even high school diplomas by colleges and employers. Reducing college bloat, middle management bloat and consultancy bloat, and ensuring that colleges and managements select on merit, should be a major long-term objective of the MAGA movement, improving life for all but drones and parasites.
-0-
(The Bear’s Lair is a weekly column that is intended to appear each Monday, an appropriately gloomy day of the week. Its rationale is that the proportion of “sell” recommendations put out by Wall Street houses remains far below that of “buy” recommendations. Accordingly, investors have an excess of positive information and very little negative information. The column thus takes the ursine view of life and the market, in the hope that it may be usefully different from what investors see elsewhere.)