Structural integrity

Recently, a former classmate in London asked if I assumed that the private sector is more flexible than the public sector.

My reply? Not necessarily- as the degree of flexibility within any organization is limited by the need to ensure what I call its “structural integrity”.

This relatively short article presents the concept and few case studies.

Structural integrity

When I first considered the concept, in the late 1980s, I was working as an “intensive number cruncher”, on a different model to crunch numbers on a daily basis.

The concept isn’t really original- it was derived from some toying with cultural anthropology- i.e. considering what is “normal” within a specific cultural context.

Formally, the number crunching models that I built or helped to build were called DSS (“Decision Support Systems”) models.

The logic was to get data and the decision making approach from the customer, and build a model that embedded that approach.

How do you do that? you have first to understand what is “structural”- or “organic”, as today we would say.

In companies, the identification of what was “organic” implied starting with either the data that was available, or the decision making process that was is in use, and test if your model produced from the same data the same results that had been produced before (a “reality check”).

If it required data that wasn’t available, or to change the decision making process, then your model had to be changed.

Unless the customer agreed to change either the data or the process- sometimes IT is used to bypass the resistance to organizational change.

Being a bookworm, beside industry-specific books, I scouted for books on decision making, creative accounting (i.e. how to trick numbers the Enron or BCCI way), data auditing and interpretation, and financial scandals due to mismanagement of information feeding decisions, like at Lloyds and others.

Working with controllers, CFOs, marketing managers, but no more than 1-2 days in a row with each one, probably I had more than a couple of dozen of “examples”, across multiple industries.

Coming from science and humanities to IT and number crunching/controlling/accounting, eventually I built a small method to formalize my iterative assert-test-improve approach to DSS- that a partner and his assistants converted into a formalized process to be proposed as an implementation to my employers’ methodology (called “Method/1”, eventually just “M/1”).

The key focus? Deliver something that is not “rejected” by the “immune system” of the organization’s culture.

The immune system of an organization

When you introduce a new process, technology, or, sometimes, just people, you have to consider how the new elements mix with the existing organization- and the available level of flexibility.

When you introduce significant organizational changes, the risk is that you unknowingly add or remove flexibility.

By reducing flexibility, you implicitly require that everything works as it should formally work.

By increasing flexibility, you create “slack” that could then be used to introduce change that has not been planned or foreseen.

An interesting example of this approach was over 20 years ago, when I was working on a management reporting for a group of companies.

I was a mere scribe, but as a “premium” I was given the wonderful task of spending the summer to assembling and creating the documentation (eventually, about 800 pages) using the company analysis tool.

A copy of the documentation had to be delivered to each of the over 50 companies that were supposed to produce the reporting, and I had to train “on site” few pilot companies.

The customer expanded by acquisition and integration of the supply chain, by absorbing former suppliers.

And the way to ensure the integration of each new company was to impose some processes (e.g. management reporting) and some oversight, by sending in as CEO loyal managers close to retirement.

The drawback? It was clearly expressed by the new CEO of one of the companies.

When he was called as his unit had not delivered the required information, he said that he meant the new position to be a kind of pre-retirement.

He said that he had asked for the information, but when the IT and other managers in the company started sandbagging, he considered that, being close to retirement, he did not want to spend his last few years in the trenches just to send some report that nobody read.

At the same time, his managers knew that he was there just for few years, and therefore had no incentive to adapt to somebody who was transient.

A typical example of “immune response” to change.

The upside is that within an organization, if the transition is properly managed, this “immune response” requires temporary monitoring- not lifelong support, as eventually the organization adapts- or the innovative element is adapted to the organization.

In both cases, the organization aims to achieve a stability that keeps the “structural integrity” of the way the organization works and how its components work together.

As a further example, consider shifting from a “gut feeling” to a “quantitative” decision making process for new activities.

The obvious positive result is avoiding potential waste, by asking managers to “prove” their case, i.e. demonstrate quantitatively that their proposed initiative contributes to the bottom line within a reasonable timeframe.

But such a shift could also impede innovation, as usually this shift requires to provide a justification based on historical data also for new investments that focus on a market niche that had not be targeted before: the new “immune response” will reject innovation.

But a similar issue happens when you remove elements, e.g. by “streamlining” processes or downsizing.

Flexibility and size

Any organization has a certain degree of flexibility, that is inversely proportional to its size.

As the organization increases in size, efficiency requires sacrificing flexibility to ensure repeatable activities delivered in larger volumes with lower cost per unit of activity.

When downsizing, often the cut is delivered across the board, and activities are externalized either to independent units or to “service providers”.

Sometimes, this externalization process transfer outside the company also the “knowledge workers” who have can evolve the activities- unintentionally reducing the company flexibility and long-term structural integrity (affecting negatively what is called “business continuity”).

An interesting case study was a small project (again, a DSS) that I had to deliver over 20 years ago, to help a controller monitor a distribution network.

In that case, the company was a new unit delivering financial products- but, instead of re-inventing the wheel, i.e. creating a distribution network, an existing one had been co-opted.

A typical win-win, as this kept the size of the unit small, and avoided the long learning curve required to create a distribution network from scratch.

With a small glitch: how do you set the rules to control something whose business you don’t know?

In the end, after a study that had delivered a large number of what we now call “KPIs”, I was called in to build with the controller a small prototype (5 days, I think).

My first step? I had quickly read the document, then asked to the controller: what do you really need? And which data is available?

So, in few days more than those originally budgeted, a small model using the data really available was built- the aim being to control the network without requiring any new information, so that the customer could monitor what was really being done by the network, instead of having them alerted and “massaging” the data.

Years later, after I had left the company, I was told by a colleague that the system was still in use and being evolved.

But the flip side of the coin, downsizing or spinning-off by splitting an organization into smaller entities does only increase the need for “communication interfacing”, if the processes are not redesigned for the new “virtual company”.

My practical suggestions (and organizational designs) varied according to the specific organizational culture, number of external suppliers involved, industry, and so on.

The “structural integrity” can happen by chance but, frankly, it is more manageable if it is by created by design.

Designing for structural integrity

I hate cookbooks on organizational design: every time I had to design a process or revise an organizational chart (or create a new one for a startup), I first tried to understand its internal “politics and culture”- and its potential interactions with the current and targeted market or other parties involved (“stakeholders”).

I saw some “cloned” organizational charts, i.e. “standard”- as the old joke goes, “horses designed by a committee”.

As an example: if you deliver a service, you want it to be efficient (lower and measurable cost per unit of activity) more than just simply able to deliver a result at whatever cost- otherwise, you would never had the economies of scale.

But if you are an organization trying to find new applications for new ideas, you cannot adopt the same method.

You have to manage the risk of losing control of your budgets, such as somebody trying to build pocket nuclear reactors such as the one used in Asimov’s books.

But you cannot expect the same kind of monitoring approach that you use to deliver time and again the same known result- you have to adapt the level of flexibility (and the controlling approaches associated) to your own specific needs.

Moreover: both have to evolve (unless you are a monopoly belonging to the state and no privatization is in sight, while 100% of your customers are local and nobody will be ever legally allowed to compete with you- I have no example in mind!).

As more than 20 years passed, and plenty of books have been published on the subject, I will summarize a short business story, about what I observed in the first company I worked for.

A case study in keeping and evolving the structural integrity.

A small case study: a multinational consulting company

My first employer was a software company, created to be a kind of “software pool” for the Italian operation of the consulting arm of Arthur Andersen.

Similar operations were at the time available in Spain and Southern America, and while we received everything from Andersen (we were even asked to sign the “Personnel Reference Binder”, a book in English telling us what we were supposed to do, derived from the auditing activities of Arthur Andersen), our company belonged to partners- formally, we were an independent unit.

Over the years, the system integration and IT consulting activities at Andersen became more and more relevant, up to the point where it became a larger slice of the turnover than auditing and tax services, the original business or Arthur Andersen.

Being a partnership, the company was structured in “vertical gardens”, i.e. each partner had a small pyramid of managers, seniors, etc.

Therefore, growing in turnover did not imply having more seats at the desk.

This generated some friction, both within Andersen Consulting (with competition between the vertical industry divisions and the technical support organization) and with Arthur Andersen.

Part of the value of Arthur Andersen was a large operation near Chicago whose purpose was to collect the knowledge produced by any activity worldwide- a kind of “knowledge base” ante litteram (official public brochures at the time said that 5000 people out of 70000 worldwide worked there).

How long did it take to produce the split between the two parts? An internal training course at Andersen stated that the company worked with IBM to produce the Customer Information Control System (a.k.a. CICS) in the 1960s, when it was considered a “fad”- as all the competitors worked on punched cards, not with screens.

I left the company in 1990, and the two split in 1989, while for the next decade kept fighting on IPR rights and related payments- part of the severance agreement included the requirement that Andersen Consulting cease to use the “Andersen” name, and an internal contest was run.

The winner? Accenture- whose internal meaning was supposed to be AC Century- i.e. Andersen Consulting Century.

Anyway- the final split occurred just in time, as soon Arthur Andersen was dissolved, after the conviction for Enron.

But, despite the separation, the “imprint” of the original shared culture (methods, training, etc) lasted longer.

I saw the first signs of divergences in the culture when the company shifted the training for the “main company” from St Charles to computer-based training and decentralized training around the world, weakening the cultural link.

Probably today it would be possible to keep the “structural integrity” of the internal culture, as people in their 20s are used to keep links via Facebook, Skype, etc.

But sometimes you need to promote, not just to keep structural integrity.

Promoting structural integrity: banking in Italy

I have been working in the banking industry since the late 1980s, mainly in Italy but also in few other European countries, and thanks also to recent events, I think that anybody reading newspapers or watching news on TV has heard of Basel II (soon Basel III), risk management, and so on.

Structural integrity is not necessarily something limited to a single entity (a company or an organization), or a “supply chain” (e.g. a major company and its suppliers, notably when you are operating under “lean” conditions to minimize or optimize costs and increase quality).

It could extend to an entire economy (e.g. how can you produce if no financing is available in your country?) or an entire industry (e.g. as shown by the temporary halting of bank-to-bank lending during the real estate crisis, due to a lack of mutual trust).

I have been lucky, as my first banking project in late 1980s was to roll-out a general ledger for a large Italian bank.

Therefore, I was able to see the connection between all the systems.

Thanks to the credibility derived from that expertise, I was involved in project across different areas of banking, from the front-office (what you see in the branch, and what they do at the end of the day), to the back office (from day closing to provisioning and settling to ensure that the bank is able to operate).

When I started, most large Italian banks belonged indirectly to the State and were structured around specialized banking entities- a side-effect both of a major scandal at the end of the XIX century and the dismantling post-WWII of the Corporative State created by the Fascist government.

Eventually, Italy expanded privatization, with the Bank of Italy at first playing a pivotal role on two issues: keeping the Italian banking industry under Italian ownership, and using “moral suasion” to have stronger players absorb weaker (or failed) ones (e.g. few smaller banks, and former issuing banks).

Due to extreme politicization, official data in the 1990s claimed that Italian banks had been 30% overstaffed, and were used to operate with a “spread” (margin between interest paid and interest received) of up to 7.5% (or 750 basis point), with an extreme fragmentation of the market (local banks had been created centuries ago as “shared financial pools”).

Both the Government and the Bank of Italy knew that eventually the European partners would require to open up the market, and therefore, beside mergers, the Bank of Italy allowed some foreign banks to enter the market (e.g. Deutsche Bank via the acquisition of the Banca d’America e d’Italia in the late 1980s), while also trying to control the systemic risk.

Major Italian companies often received financing with no economic relationship between the risk and the benefit (e.g. the Ferruzzi Group in the late 1980s, or Parmalat in early 2000s, to quote some of most recent cases).

In each case, the Bank of Italy and the Italian Bankers’ Association increased the level of reporting on risk that each bank was required to provide.

From the prevention of systemic risk, to group risk and exposure via third parties, to an oversight of the operational costs.

Eventually, the Italian Parliament expanded the “coverage” to any institution involved in risk.

Structural integrity and systemic risk

I hold an Italian passport, and I was born in Italy.

My Eastern European friends in mid- to late-1990s said that until then, they had met only Italian who either approved or criticized Italy- while instead I said that Italy had some positive albeit unknown aspects.

On the issue of risk management, from late 1987 until 2007 I saw a growing expansion of the reporting requirements, expanding from the banking industry, to any financial entity.

The “structural integrity” in this case is built via the creation of shared communication processes, and enforcing strict controls on data quality.

A balancing act is required, to avoid the disclosure of competitive or sensitive information.

When, years ago, I helped to negotiate an agreement (not completed) to add a risk management user interface based on a factoring risk management product built in UK for an Italian customer, the first issue was to study and explain the significant differences in reporting requirements between UK and Italy.

It is more detailed- but, to avoid a lengthy dissertation: if you had a new potential customer, you could check the exposure in the system, without knowing the details of the exposure, and pay a fee; for your existing customers, you would keep receiving information about their financial commitments within Italy.

Most Continental Europe countries have a similar “centralized risk monitoring” system, but usually limited to the banking industry.

Recent events, ranging from the real estate crisis to the new precursor of the EMF (i.e. the EU version of the IMF) created a general consensus in Continental Europe on the need to improve the preventive risk management, by increasing the information available.

Beside “reporting” the existing risk, the structural integrity of the system can be ensured by using the data provided in a predictive way.

Eurostat and the National Statistical offices can collect information about the market evolution by industry, while the Central Banks can collect information about the exposure within each territory and industry.

Instead of extinguishing “financial fires”, this information could be used to indirectly or directly influence the intervention- as suggested for the new Basel III.

If you want: the flexibility can be left in the market, provided that this does not affect the structural integrity- moreover, if we consider the demographic evolution of EU.


One thought on “Structural integrity

  1. Pingback: On methodologies – from Method/1 to Agile | GettingAroundTheWorld.Net

Leave a Reply

Please log in using one of these methods to post your comment: 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