the double life of brian
consultants, once more, and the world that could have been
Welcome, FTAlphaville readers! (I must remember to pitch them more often). While you’re here, why not pre-order a copy of “The Unaccountability Machine”, a book about big decision-making systems and how they inevitably go bonkers. I have quite a bit about management consultants and their foibles in that book, so I’ll continue on from the previous post here, asking - is the Old Rectory Mittelstand as good as it looks?
Because the question that struck me, shortly after publication, and then again when Louis wrote about the subject for Alphaville, was - am I falling into a fallacy here, assuming that because something looks big and important in the statistics, that it must consequently be a strength rather than a weakness? Am I once more in the mental prison of the accounting system?
The question here would be - consider a consultant, called “Brian” (thanks to Alex Harrowell). He works out of the Old Rectory somewhere, and there’s not much that he doesn’t know about embedded software systems for monitoring dry powder flows in the chemical manufacturing industry. Back in the 80s, he was on the graduate trainee scheme at ICI, but one thing led to another and now he travels the world helping people set up, maintain and troubleshoot. He has never been asked whether his invoices should be characterised as “Management consultancy” or as “Technical, trade-related and other business services” and would probably just flip a coin if he was.
Brian - and Dennis, and Hitesh, and Pamela and all the other small consultancies working out of decommissioned ecclesiastical accommodation up and down this great land - makes a good living, and is a solid contributor to our balance of trade. But … is this perhaps the reason that ICI plc isn’t such a big thing any more? Could there have been an alternative history, one in which Brian, Dennis, Hitesh, Pamela and all had stayed in the corporations and government departments where they learned their trades, and made their contributions to marginal and not-so-marginal productivity improvements there?
I absolutely, one hundred per cent do not know the answer to this. But a few things covered in the book make me nervous.
In particular, cybernetics is all about information flow – a great part of the system is about ensuring that information gets from where it is, to where it needs to be, in time and in a form in which it can be the basis for action. (This is why it’s dumb, IMO, to complain that a consultant has just “talked to the employees who already knew the solution, then dressed it up in nice language for the management” – all this means is “the company had a management problem, then the consultant came in and solved the management problem for them”).
The problem of information transfer is always at its most acute when communication needs to cross a boundary between decision-making units. My guy Stafford Beer called this the problem of “translation and transduction” (transduction being, for example, what happens when light hitting the retina causes chemical-electric changes in the nerves, preserving the information content).
Whenever you substitute a consultant for an employee, you’re substituting an internal boundary for an external one, and most likely making the problem of translation and transduction more difficult. You have less bandwidth in the communication channel, not least because the kind of things you can communicate with a consultant are partly circumscribed by the engagement letter, a document which has usually been written precisely because you didn’t understand the problem.
So … the “consultantisation” of the British industrial system might definitely be a problem, which might specifically show up as a productivity problem very similar to the one we can actually see in the data (concentrated in smaller companies, associated with slow take-up of best practice and new technologies, etc). In the book, I advance the idea that outsourcing in the public sector has been a massive source of this sort of problem.
The other thing that worries me about Brian is that he might be part of another known problem of British industry – reluctance to invest in training. If you suspect that your most skilled technicians and managers are going to leave and go into consultancy half way through their career, you’re going to be that much less willing to spend money training them, and consequently more inclined to complain that the government and universities aren’t doing enough. Again, I would not invest many credibility points in this, but it does sort of match the fact pattern.
Which leaves the final and most worrying question of all – what if Brian, Pamela, Hitesh and Dennis are all symptoms of the underlying malaise. Britain is special and unique in this sector, hurray. But … one might also ask, “what is it about British industry and government that makes talented specialists and managers decide that they have to leave it and set up their own consultancies”?
I’ll return to that, on Friday, in about the most trivial and silly way possible I hope.
I think the "why so many consultancies" has at least a couple of drivers - one being the pension and tax/NI system, where an experienced specialist can suddenly become very expensive indeed to have on the books. So cut 'em a deal on early retirement, and bring them back as consultants/contractors. And then maybe something, something, VAT.
This implies that a serious slug of consultancy firms are one-man bands, micro businesses, with the owner/operator being in their fifties/sixties. May be older, if they're really enjoying it.
The other driver would be regulatory or technological changes. There'd be a set of employees who would have a first mover advantage, and be able to rake it in due to a demand/supply imbalance by leaving the employer and setting up on their own as soon as. Or, there's MiFiD II and research houses.
This gives a set where the owner strikes out alone around the age of 30-ish. Likely sectors would be HR including recruitment, IT, marketing, accounting, general management. Probably still one-man bands, geographically limited.
However, at a guess, that second set is more likely to shift from micro to small, and probably survive longer, and shift out of the Old Rectory to Ye Olde Bank Branch.
If there is a problem, then the first group can be solved - dick about with the tax system, job done. Ahem.
The second group, you might not be able to do that much about - save for wandering into Niven's Gil the ARM territory, or Herbert's McKie & BuSab. Tricky, but see the whole AI regulations issue. Then again, let us not go there right now.
Then there's possibly a third group - which might be more interesting, in that it might hint at a structural issue. It's also annoying, as it's beginning to look as if I might actually have to pony up some cash for the book, at the very least.
Anyway, it boils down to "I don't know, more research needed, maybe it'll go critical tomorrow", but I suspect your cybernetics/information processing model doesn't fit. It's an OCP.
Can't remember the name of the book offhand, but the ascendance of Silicon Valley has been blamed on California's refusal to enforce no-compete clauses. This creates precisely this outsourcing problem which you discuss. I suppose it is all contingent: do the higher-powered incentives of individual entrepreneurship outweigh the informational advantages of concentration?
I'd say the public-sector problem is different. The public sector simply does not want to pay enough, at least in the US. And it is also more bound by internal fairness norms, which conflict with market pay norms. The famous public-sector discount is real enough, but affects different kinds of workers differently. A lawyer or economist might take a bigger haircut because the work is more fun than private-sector work, but an IT person might not see things the same way.