The perils of the “Client-Server” outsourcing business model – a.k.a “don’t let bean counters run your business”

One of my hobbies – i.e. stuff I do exclusively for fun, not primarily for making a buck – is to drive bus once-in-a-while within the public transportation system of Stockholm.

Why… Someone might ask….?

Because it’s  great fun and very rewarding to drive an 18 meter long bus, taking up to 100 passengers,  on the busy and narrow streets and roads of the Stockholm area, where the “margins of error” when driving such a very large vehicle can be counted in centimetres and fractions of seconds – it takes a lot of focus, skill, situational awareness and excellent driving technique to maneuver such a vehicle safely, timely, comfortably and environmentally friendly in such an environment. Secondly, driving bus in the public transportation system gives me an opportunity to encounters and interaction with a very different strata of people from  very different walks of life that I otherwise would never encounter, not in my profession, nor in my private life. Third, I like the immediate “feedback loop”: in the bus driving business you will immediately see the results of your efforts: either you are on time, with no damage (physical or psychological) to the passengers, third parties, environment nor the vehicle, or you are not. The feedback is very binary and very immediate, as opposed to my profession, where results and outcomes might takes months and even years to manifest themselves,  and way too often are far from being objectively concrete and crisp (except in purely financial terms..)

Anyway, let’s get to business:

Since many years, the public transportation business in Stockholm has been fully commercialized, based on a “client-server” model,  where the Stockholm County’s public transportation council  is the “client”, defining the requirements and penalty clauses by issuing RFP’s,  and a number of bus companies – the “servers” –  compete for the business in a traditional RFP-process.  Thus, the model is a traditional “business-by-contracts” model, with very limited stakeholder involvement. In other words, the business decisions are made almost exclusively on strict economic/financial grounds, with cost (i.e. the price tag on the response to the RFP)  being the by far most important item on the proposal. Thus, the focus in evaluation of the various RFP’s is almost exclusively on the “cost”, i.e. the price tag, whereas the needs and value proposition for everyone else, including the “end users” – i.e. the public & passengers – as well as the employees,  takes the back seat, very very far from focus.

Thus, the public transportation system business is exclusively managed and run by “empty suits”, i.e by folks with very limited or no skin in the game (the politicians populating the traffic council do not commute by bus or by any other public transportation) and they have no personal responsibility, knowledge nor interest in the business, over and above the cost dimension.  Basically, in order for these empty suits to “look good”, they only have to look at the cost/efficiency side of the equation, the value side is not relevant.

Recently, the RFP-process for the region of Northern Stockholm (where I live) was won by Arriva, having won by placing a bid with a price tag approximately 30% lower than the “second best” competitor. 30%!!!  And this RPF  was for an already established network of public transportation, meaning that Arriva’s proposal was based on a belief that they could take over and  operate – and make profit from! – an existing business where the previous entrepreneur had already been struggling to make both ends meet in financial as well as operational terms. In other words, Arriva got the business by committing to the public transportation council that they would be capable of operating the existing traffic volume – which the previous operator barely had managed to do – with a 30% cost reduction….!  

A 30% cost reduction in any established business is of such a magnitude that it will get any financial bean counter to wet his or her’s pants by pure euforia, and the capability for logical thought of these financial bean counters – if they actually have any – will vanish completely when facing “such a fantastic business proposal”!

Needless to say, Arriva proved to be absolutely/totally/beyond any doubt incapable of delivering: they have now had the operations going for 5 weeks, absolute chaos reins in the Northern Stockholm public transportation system,  and it has become painfully obvious to anyone involved that the “fantastic” business proposal, with 30% lower price tag, was a pure fantasy, only able to fulfill the wet dreams of the empty suit bean counters on both sides of the contract,  but with no connection what so ever to the reality.  Busses are not running, they are cancelled, they are late, they run out of fuel, they are not ready for traffic in the morning, they get engine or other failures at a previously unseen rate, drivers are leaving Arriva en masse, and the passengers are getting desperate trying to figure out alternative ways to get to their work, schools or wherever they were planning to go. The only business gains made since Arriva took over January 7th has been made by the car sales businesses – people in Northern Stockholm buy cars at an unprecendented rate since January 7.

Basically, it appears that all the stakeholders of the Northern Stockholm public transportation system have been hit by a Black Swan!  But in fact, in this case  it’s not a Black Swan that hit, since this mess could have been predicted by anyone with the tiniest knowledge of complex systems, stochastic processes and randomness, in the case of the bus business in Stockholm  it was the total incompetence of all de involved decision makers that caused an otherwise preventable catastrophe!

Let’s look at the root causes for this mess:

First, the business was under a lot of financial and efficiency pressure already before Arriva won the new contract. It’s hardly so that the previous entrepreneur, Keolis, was making huge profits from the business, in fact, the Keolis shareholders would have been better off if Keolis had ceased to operate the bus business and placed their investments in a normal bank account. Thus, there was from start absolutely no redundancy or fat to be eliminated to reduce “excessive” margins, or operational “ineffeciencies”. The system was already running at its limit, financially as well as operationally.   Secondly, in the bus business most of the significant costs are fixed, regardless which entrepreneur or operator runs the business: buses tend to cost the same, fuel tends to cost the same, and the costs for staff is the same, regardless of operator. So where did Arriva think they would find the 30%  “fat” to cut from the system….?

The answer is “utilization factor“.  That is, in the bus business there are two key resources that determine the “efficiency” of the system, namely the vehicles and the drivers. For a simplistic, economically trained mind, such as that of the typical MBA-type empty suit, it is very tempting- in the name of efficiency and cost reduction –  to maximize utilization of the key resources of the system, which in the bus business means to mazimize the utilization of the vehicles (busses) as well as of the drivers.  To mazimize utilization, the only thing you need to do is to make sure the key resources are as busy as possible, with no “idle time” or slack what so ever. Idle time/slack in the bus business translates immediately to reserve capacity and managed redundancy to deal with the unplanned. By eliminating all reserve capacity, e.g. stand-by busses and drivers, and eliminating all slack (reserve time)  in the route timetable,  you can drive the utilization factor, and thus the “efficiency” of the system, to unprecendented levels. And that’s exactly what Arriva’s bean counters have done: they have used their financial engineering knives to carve out all “slack”, i.e. spare capacity, from the system, relying fully upon their deterministic planning, where “unforseen” events remain absent, and where not a single cog in the system will ever fail.   Unfortunately, this type of “financial engineering” completely misses the lessons available in basic queuing theory, where it’s painfully easy to demonstrate that any system –  particularly a non-linear dynamic system such as a transportation network, where every discrete temporal system state impacts the next –  with even the tiniest randomness involved (“stochastic processes”)   where you start pushing the utilization factor upwards will soon have its overall performance (“throughput”) completely killed due to congestion and delays caused by the high utilization factor.   And the bus business is full of random events, e.g. traffic events, as well as of positive feedback loops  capable of pushing the overall system into a chaotic state that can severely disturb the deterministic plans, predictions and financial forecasts of the financial empty suits.

Take a look at the graph below: it’s a very simple mathematic model from queing theory, demonstrating the effects on overall system response time when the utilization factor of any key resource is driven upwards:


[ Little’s Law: W=1/(μ-λ) ]

Whenever utilization of any key resource starts to exceed some 75%, bad things start to happen to overall system throughput.  For instance, at 80% of utilization, the overall throughput time (the time you spend in line waiting to be served + the time for the actual service)  is four times the service time, and anything over 80% utilization will quickly lead to total collapse of system response.

Unfortunately, this type of thinking does not exist even on the long range radar screens of the financial bean counters of our modern “efficiency based” shareholder focused economy run by empty suits with an MBA, who run the business by check lists and efficiency metrics exclusively, and the results of this type of thinking are painfully obvious to anyone who is depending on the value side – not only the cost/efficiency side – of the business.


About swdevperestroika

High tech industry veteran, avid hacker reluctantly transformed to mgmt consultant.
This entry was posted in Business, Complex Systems, Leadership, Management, Organization, Systems and tagged , , , , , , , . Bookmark the permalink.

One Response to The perils of the “Client-Server” outsourcing business model – a.k.a “don’t let bean counters run your business”

  1. Jonas says:

    Nice chant. I’d guess that one of the reason is not that the “empty suites” holds an MBA, i’d say that the problem rather might be that the “empty suites doesn’t hold an MBA.
    Education is good, one might even learn Little’s law at an MBA.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s