Complexity-based Identification of Defaulting Companies


The search for early warning signs is one of the key issues for decision makers, managers and investors. The advantages of knowing in advance the evolution towards critical situations are obvious.

Complexity is a new and powerful indicator that quantifies the degree of sophistication and governability of a business and which impacts its Resistance to Shocks (RtS). Both complexity and  RtS indicators establish a radically innovative means of anticipating crises. As of today, conventional methods have failed to identify such signals, especially with this level of detail.

Through the Quantitative Complexity Management (QCM)  any business which is undergoing severe stress or is on a path to collapse shows a rapid complexity fluctuations. A consistent and continuous reduction of complexity is a typical pre-alarm. If this occurs, then a decision maker, e.g. a trader, is able to put in place all those actions needed to avoid financial losses and/or take advantage of these situations e.g. via shorting.

Fact:  Changes in complexity indicate imminent crisis.

In the next paragraphs it will be shown how Complexity and Resistance to Shocks (RtS) of Balance Sheets are able to provide early warning signals of an imminent crisis.

We have analysed the complexity of 3 listed companies whereby:

  • At the end of July 2016, two companies filed for bankruptcy and
  • Since August 2016, one company operates under judicial management

The bottom line is that these crises could have been be predicted with considerable anticipation.

Based on the measurement of complexity the first symptoms have been recorded as follows:

  • Swiber Holdings: from the 1st quarter 2014, i.e. two years before judicial management.
  • Halcon Resources Corp: since the 2nd quarter 2014, i.e. seven quarters prior to default
  • Atlas Resource Partners: from the 2nd quarter 2015, i.e. three quarters prior to default

The plots below compare the evolution of Balance Sheet complexity versus Total Assets of the three companies.

In all cases Total Assets of each Company were still growing, while situation of concealed crisis have been revealed by the corresponding complexity trend.

The consistent complexity decrease in all cases points to an increase of ‘de-correlation’ of the corresponding Balance Sheets. This means that the relationships between the various Balance Sheet entries are becoming more independent (i.e. the correlations are becoming weaker, pointing to a ‘less structured’ situation).  Consequently, the business is more ‘disordered’, hence difficult to control/predict and is therefore more exposed to unexpected events.

Let us examine the three cases:

Swiber Holdings  

The growth of the company’s Total Assets, between the 4th quarter 2013 and the 2nd quarter 2014, conceals a critical situation.

A complexity reduction of 20% of within just one quarter is a typical symptom that something is going wrong.  The compounded quarterly growth rate of complexity (-4,1%), double of that of Total Assets (-2,0%). Such complexity decrease is mainly due to growing disorder (lower correlations) in the Balance Sheet. Basically, the structure of the information within the balance sheet is becoming weaker, pointing to increased exposure.


Figure 1. Complexity and Total Asset evolution of Swiber Holdings (Q4 2013=100)

Halcon Resources Corp

The growth of Total Assets between the 4th quarter 2013 and the 4th quarter 2014, hides a critical situation highlighted by complexity reduction: -23% compared to a growth of  Total Assets of  20%.  As in the previous case, the Balance Sheet is becoming more ‘disordered’, pointing to a more fragile business.


Figure 2. Complexity and Total asset evolution of Halcon Resources Corp (Q4 2013=100)

Atlas Resource Partners

The contained decline of its Total Assets, between the 4th quarter 2014 and the 1st quarter 2016 (-5%), hides a critical situation reflected by a complexity reduction of -24%. The situation is becoming critical from the 2nd quarter 2015 onwards. The -24% of complexity reduction within a half-year period is a typical symptom of a failing/declining system.


Figure 3. Complexity vs Total assets evolution of Atlas Resource Partners (Q4 2014=100)

In conclusion, the monitoring of complexity provides businesses with new information of systemic nature which reflects the interplay of structure and disorder in Balance Sheets, and which may be expanded to include Cash Flow, Income Statements or Ratios. A sustained loss of complexity points to de-correlation which, in turn, reveals a growing level of disorder. This constitutes a formidable early warning mechanism which classical Balance Sheet analysis will not reveal.

The above analysis provides also the complexity drivers responsible for loss of complexity. This information, which is crucial in order to understand better a situation of crisis and fragility, has not been presented for reasons of brevity.

And now a look at AIG and the associated scandal (September 16, 2008). From a 2010 article:

“The government’s $182 billion bailout of insurance giant AIG should be seen as the Rosetta Stone for understanding the financial crisis and its costly aftermath. The story of American International Group explains the larger catastrophe not because this was the biggest corporate bailout in history but because AIG’s collapse and subsequent rescue involved nearly all the critical elements, including delusion and deception….”


“Bailing out AIG effectively meant rescuing Goldman Sachs, Morgan Stanley, Bank of America and Merrill Lynch (as well as a dozen of European banks) from huge losses. Those financial institutions played the derivatives game with AIG, the esoteric practice of placing financial bets on future events. AIG lost its bets, which led to its collapse. But other gamblers—the counterparties in AIG’s derivative deals—were made whole on their bets, paid off 100 cents on the dollar. Taxpayers got stuck with the bill.”

The importance of knowing that a company is manipulating its books cannot be overstated. Going through the wreckage is far worse than running a simple quarterly check.

From the analysis of the above companies we have seen that:

IF the complexity trend of the balance sheet is negative
The “Total Assets” trend is positive
The company is moving rapidly towards an imminent and serious crisis

We will show, how based on this simple rule:

the AIG default could have been predicted in September 2006, a year before it happened

First of all, let us examine the evolution of AIG’s complexity and Total Assets:


The +20% increase of AIG’s Total Assets between Q1 2006 and the Q2 2008, hides a critical situation. Two key issues which would have pre-alarmed decision makers (FED?) about  the state of health of AIG’s busines:

1.  The systematic and continuous decline of the complexity which begins in Q1 2006. Basically, the structure of the data in the balance sheet’s is becoming increasingly weak. The rules and the information available to manage the business are increasingly fragile. Therefore, the exposition to unexpected events is greater.

2. The rapid decline of complexity starts in Q1 2006, followed by a second massive dip of -33% in only two quarters. It would seem that the attempt to restructure the AIG business during the Q2-Q4 2007 period has produced little results.

The bottom line is: the first symptoms of AIG’s crisis were evident since Q3 2006, i.e. a year before the U.S. government  bailed out AIG with $85 billion .

The consistent complexity reduction indicates an increase of ‘de-correlation’ of the corresponding Balance Sheet items. This means that the interdependencies between the balance sheet entries are becoming weaker (fuzzier). Therefore, the business is less controllable, predictable and governable. This means that it is more exposed to unexpected and potentially destabilizing events.

However, there is more than just governability at stake. When a business becomes ‘fuzzier’, it becomes easier to engage in fraudulent activity, to hide inefficiencies and incompetence. The transformation of structure to entropy is rarely a good thing.

More soon.

QCM methodology and the related software tools allow to quantity the complexity and the RtS. The QCM approach complies  with UNI 11613 Business Complexity Assessment guidelines. UNI 11613 furnishes guidelines as to how an organization may establish, monitor and put into practice the assessment of the complexity of its own business. The goal is to identify the critical complexity of a business and to identify and rank its drivers, with the objective of increasing the resilience of an organization.

Resistance to Shocks (RtS), sometimes known as resilience, or ‘shock-worthiness’, measures the capacity to absorb shocks or destabilizing events, such as financial contagion, stock market collapses, market bubbles, natural disasters or geopolitical events. It provides an indication of how stable a company, portfolio or market is and how it will react to the said events.


One thought on “Complexity-based Identification of Defaulting Companies

  1. Are there any situations where an unexplained rise in current complexity can also be a predictor of financial trouble for a company. In particular, if a company seeks a much wider range of customers in order to maintain sales “at any cost” it is likely that the complexity of the company’s operations will increase. However I understand that this may not affect the Balance Sheet.


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