Ant colonies, self-organized criticality and small-caps

18 September, 2020

A lot has already been written about the recent SEBI order on multi-cap mutual funds (announced on 11th Sept); and we do not intend to belabour the topic further. To us, the market reaction to this directive is far more interesting. It brings us to examine as to how the collective decision-making works in ‘complex adaptive systems’ and how our investment strategy should reflect that. The complexity at work here is: (a) variables for the fund house: merge scheme, or buy and comply, (b) variables for SEBI: extend time to comply, allow flexi cap as a new category, (c) variables for retail/HNI: pre-empt mutual funds and buy now, or sell on possible rallies. We use the analogy of ant-colonies to see how the system adapts to it.

Ant colonies: Ant colonies demonstrate a fascinating example of the concept of self-organization—a process where a structure appears in the system that does not have a central authority that imposes its will by pre-planning (and we submit that markets operate in a similar fashion, the recent overpowering role of respective central banks, notwithstanding). When ants go foraging for food, they initially spread out in many different directions; in a pretty random manner. However, once they locate the food, they return to the nest while laying down the pheromone trail. Whenever one ant finds the shortest path, its quicker return and multiple trips intensifies the concentration of pheromones along that path. Other ants, then, choose the path with the strongest concentration of pheromones, thereby solidifying the path further. This means that a few ants end up working more than others, but their behaviour is directed to the survival of the colony, rather than the survival of any individual ant. This is how the collective behaviour of an ant colony finds an optimal solution, to a very basic problem.

Since the SEBI order, the BSE Small cap index is up 5.1% in one week versus 0.5% returned by the Sensex. This is despite the largest mutual fund (in the impacted category) publicly announcing that they are averse to buying the small cap stocks just to meet the requirements in this order. To us, this is the markets’ collective behaviour finding an optimal solution, and while important, SEBI’s order is not solely responsible for such market behaviour. The example below will clarify further.

Self-organized criticality: Whenever a large-scale event occurs, it is a human tendency to gravitate towards one specific, easily identifiable cause, as being the sole responsible factor. However, numerous scientists have pointed out that large-scale events are not necessarily the result of a single large event, but rather the unfolding of many smaller events that create an avalanche-like effect. Per Bak, a Danish theoretical physicist, explained this beautifully, as he developed the ‘self-organized criticality’—a holistic theory of how such systems behave.

Imagine an apparatus that drops a single grain of sand on a large flat table. Initially, sand spreads across the table and begins to form a slight pile. As the grains rest, one on top of another, the pile starts rising—forming a slope on either side. Eventually, the pile of sand cannot grow any higher. At this point, the sand starts trickling down the slope at a faster pace than the pace at which the grains are added on the top, creating sort of an avalanche. At its highest level, the sand pile is in a state of criticality—on the verge of becoming unstable. The last grain did not create the avalanche; the system was unstable, in a state of criticality and waiting for trigger for the eventuality (avalanche, in this case) to manifest.

Small cap Index: Over the last seventeen years through March-2020, the CAGR in returns of Sensex, Midcap and Small cap indices are 14.5%, 15.3% and 15.0% respectively—there is hardly anything to choose between them. And yet, as the table below shows, the difference in annual returns has been extremely stark.

Over the last two years, the markets seemed to have gravitated towards the theory that ‘a good company is a great investment at ANY price’. Whereas, well-managed companies, operating in sectors that have sizeable moats, are certainly good businesses to own; but for us, for them to be great investments that generate superlative returns on a cross-cycle basis, the price paid must be right as well. Historically, we have witnessed great companies delivering near ZERO returns when the initial price paid was simply too high (Coca-Cola, 1998-2016, EPS doubled during that time; IBM, 1999-2010, EPS tripled during that time; Hindustan Unilever, 1999-2010, EPS doubled; Colgate 1994-2009, EPS quadrupled).

Between Jan-2018 and the SEBI order, the small-cap index had underperformed the Sensex by c30%. In the past, indices have mean-reverted based on their earnings and valuation cycles. The SEBI order, then, was just the last grain in the metaphorical sand-pile of the market which had reached a self-organized criticality, and the collective behaviour of the market participants led to the outcome that we saw during the last week. It has happened previously, and we will not be surprised if it recurs. As we have previously written—whatever is intrinsically unsustainable, will find out a way to not sustain.

2 thoughts on “Ant colonies, self-organized criticality and small-caps

Ved Vyas4 months ago,

Excellent write up. We seemed to be going back to 1960s where people were willing to any price for growth company in USA. At present top – 5 tech companies in US correspond to 20-25% of S&P. One car company with no profit to show has a valuation which is great than sum of all the large car manufacturer in US.


Diwakar4 months ago,

Hey, thanks for writing, the 16 Yrs shows that equal times small cap & now in 2 Sensex have rallied & nows the time for the small category to take the lead as its almost 2 Yrs plus of they being laggards also interest rate where it is an looking forward for few rate cuts / holding on the low rate will help this category.


Leave a Reply

Your email address will not be published. Required fields are marked *