Letter # 45

Imagine you had a magic wand, which gave you the amazing power to time the market with complete accuracy, but with a small glitch—you can only trade indices (not individual stocks) and must always stay invested. Over the past two decades, how much do you think the magic wand would have been worth? i.e. how high would the returns have been had you sold at top and bought at bottom? For context, the Sensex has returned 14% CAGR and BSE Small-cap index 16% CAGR since 2003. A brief detour before we return to answer this.

Randall Munroe studied until the age of 22; first at a high school that specialized in mathematics and science, and graduated with a degree in physics. It was like a dream come true when Langley Research Centre, NASA, hired him as a programmer and roboticist. However, in 2006, NASA ‘ran out of money’ ⁽¹⁾ to rehire him and had to let him go within a year.

On the basis on what I wrote above, would you be surprised to read this bio of Randall in Wikipedia: Randall Munroe is an American cartoonist, author, and engineer.” That’s right… 15 years later, he is a “cartoonist” first, “author” later, and the education that landed him the NASA job, last!

Randall went on to write an extremely successful blog titled xkcd (not an acronym) that bore the tagline ‘A webcomic of romance, sarcasm, math and language’. His webcomic had already started garnering 70 million hits a month by October 2007.

He also wrote a blog titled ‘what if’, where he answered totally absurd questions regarding math and physics. Questions like, ‘what if… the earth stopped spinning, but the atmosphere retained its velocity’ or ‘what if… I took a swim in a typical spent nuclear fuel pool; how long can I safely stay at the surface?’ or ‘what if… everyone on earth aimed a laser pointer at moon at the same time, would it change its colour?’

While the questions are absurd, Randall’s answers are well researched, nuanced, and profound. In 2014, he published a collection of some of the questions in a book titled ‘What if’, which reached the top of New York Times bestsellers list within a month and got featured as the ‘Amazon Best Book of the Month’. Randall also has a tremendous sense of humour, which makes his book an informative as well as a fun read for people of all ages. Do give it a shot.

I am trying to follow in Randall’s footsteps by taking up random questions in finance that can be answered with tons of data.

And, the answer to the magic wand question is, a whopping 31.4%! Yes, that is what the magic wand was worth. You would have invested in the Sensex between December 2003 and June 2004, December 2007 and February 2009, October 2010 and December 2011, and December 2017 and March 2020 (and long Small-cap index all the other times).

The elephant in the room however is…we don’t really have a magic wand, do we? But we have the next best thing–the benefit of hindsight.

So, let’s formulate this. We start with buying one index (say small cap) and hold it till it outperforms the other (Sensex) by certain percentage points. Once it does, we swap (sell small cap and buy Sensex), and repeat. We need to optimise for just one variable–that outperformance number at which we will flip and switch.

Assuming we switched after 20% outperformance, the strategy would have generated 18% CAGR– higher than individual returns of Sensex or the Small-cap index, but a far cry from the ‘magic wand’ number. The chart below gives the returns that one would have generated under different thresholds of outperformance. As we can see, the closest we could get to the magic wand number is by switching out when the outperformance hits 80% (i.e., we do not sell small cap until it outperforms Sensex by 80%, and vice-versa).

This exercise highlights a few important aspects: (a) The market operates in cycles. There are times when large caps outperform mid and small caps, and vice-versa. Over the past 2 decades, there have been 9 distinct cycles of reversals. While we are in a cycle, it appears as if things will likely never reverse (small-cap cycle now or large-cap cycle between December 2017 and March 2020), but historically, the cycle has always, invariably, and inevitably, reversed.

(b) For a portfolio to outperform over a longer period, a judicious mix of large, mid and small cap businesses is essential. Even with a suboptimal switch (after 20% outperformance), returns have been higher than either the Sensex or the Small-cap index. Our recency bias prevents us from noticing the cyclical nature of the ‘trend’ that is ongoing for a few years – say between December 2017 and March 2020, the small-cap index underperformed ferociously (down 50% vs. 13% for Sensex). So marketing geniuses came up with heuristics like ‘there are only 20-25 investible stocks in India’, ‘buy strong growth, high ROE large caps, and they will generate strong returns forever’ etc. Even professional money managers had given in, when large caps formed more than 75% of allocation in many multi-cap mutual funds (90% in a few cases).

(c) At the cost of repetition from my previous letters, there are no fixed formulae that allow one to generate superior long-term returns. A cycle has reversed within 5 months (December 2003-May 2004) and has, at times, sustained for years (May 2004 to December 2007). An upcycle has generated 70% outperformance (April 2003 to December 2003) as well as 330% outperformance (May 2004 to December 2007). A down cycle has generated 9% underperformance (December 2003-May 2004) as well as 37% underperformance (December 2017-March 2020).

The idea behind this exercise is one, to have fun with data (like Randall does) and two, to be aware that like the previous cycle (of large caps outperforming) has not lasted, the current cycle (of small caps outperforming) will also eventually change. I would have told you when, but then…I am missing my magic wand!

PS: I intend to make ‘What IF’ a routine part of my writing. Do write back to me if you have some crazy fun financial questions that can be answered only if you have tons of data (we already do!). Also, remember the model where switching at 80% outperformance generated the highest returns? In that cycle, today, we are already at 60% outperformance by small caps!

Notes:
(1) Many news things, some overdue – xkcd
This letter was originally published here: Small caps outperforming Large caps big time; what does history suggest? – cnbctv18.com

Disclaimers:
Information in this letter is not intended to be, nor should it be construed as investment, tax or legal advice, or an offer to sell, or a solicitation of any offer to make investments with Buoyant Capital. Prospective investors should rely solely on Disclosure Document filed with SEBI. Any description involving investment examples, statistical analysis or investment strategies are provided for illustration purposes only – and will not apply in all situations and may be changed at the discretion of principal officer. Certain information has been provided and/or based on third-party sources and although believed to be reliable, has not been independently verified; the investment managers make no express warranty as to its completeness or accuracy, nor can it accept responsibility for errors appearing herein.