Blog

Systems Investing

Dalal Street Rules – Years 2015 and 2016

Never send a human to do a machine’s job

Agent Smith, The Matrix

Aks capital was having the best of times. Ajay Kaul was meanwhile going through a experience which was causing him immense internal dissonance. It had been four years since they had started the live testing of MIDAS which was primarily a factor fund, but with higher rebalancing frequency than most. They also used certain quantitative parameters to enter and exit. MIDAS and MIDAS-R were kicking the benchmark’s a$# in up and down years. And it started to make him wonder if he really needed a large team of research analysts for the discretionary fund. Should he just shut that fund down and move over to the other side. Or should he just quit the experiment. The discretionary fund was also doing well for investors. The base on investors however, for both approaches was different.

Luck or Skill

Ajay pondered along with his colleagues, whether they had just got lucky. The research on factors like value, momentum, volatility had been done in developed markets. And while Aks Capital had done their own research before deploying the system, doubts lingered in his and the team’s mind. The younger team members however, were more in sync with quant investing. They did not carry the baggage of previous beliefs, and neither had many of them been in the trenches in 2008. They were all members of the post 2008 school of investing.

Factor Performance In Indian Markets

Ajay started looking for research on factors in the Indian markets. And he came across a gold mine of research. No less than Prof Jayant Varma (Finance Prof at IIMA, Ex-SEBI Board Member and known as a “finance stud” among his students) along with his colleagues had done research on Fama-French Factors and their performance in India. We have covered the fama-french factors in our earlier blog – Going Quant.

Four factor model in Indian equities market was published by Professors Agarwalla, Jacob and Varma in September 2013. Not only that. They maintained a site with the original dataset and updated data every month. The data for their research was provided by CMIE and the study was conducted for the period 1993-2013. This period was chosen as interest rates were deregulated only in the early 1990s. The study was free of survivorship bias as the Prowess database maintained by CMIE took care of this issue. This was key. Usually studies would take a universe of existing entities and do an extensive backtest or research on them. This paper provided both sets of calculation – with and without survivorship bias adjustment. It was found that survivorship had little impact on historical factor returns.

Factor Premiums in India
(Source: Four Factor Model in Indian Equities Market (2013) Agarwalla, Jacob, Varma

There were some outcomes which were in line with global performance but the most surprising one was – wait for it – SIZE HAS A NEGATIVE PREMIUM. Long thought to be a treasure trove for stock pickers – small caps did not out perform large caps. The other conclusion that the the paper came to was that “momentum earns significant positive returns in the Indian market“. This was a lot of validation for Aks’ models that were built on momentum as a core factor.

There was and still is the notion that momentum = punting = trading = gambling and so on. Most professional investors in private will confirm that price is an important input in their decision making. But never in public. Most of them model themselves as being stoic about price movements and when questioned what do they plan to do if their stock is down 50%, couple of Warren Buffett quotes will be used as defence, such as, “Buy as if the stock exchange could be shut for 10 years”, “Be an owner, not a renter of stocks” etc. But what these investors fail to emulate is that their hero is bought quality stocks, and if he made mistakes like IBM, he got rid of it and had no shame in selling it at a small loss. He was never in love with his stocks. He made it sound that he was, but make no mistake, he never is or was in love with his stocks (Ok Coke maybe). Buffett has often said that an investor also has to learn “to think about market prices“. But unfortunately this statement is never written about as his other quotes. Anyways, enough of WB and back to Aks.

Performance 2015

Performance Table

For MIDAS the year 2015 ended flat. It failed to beat the mid-cap index. MIDAR-R on the other hand trounced the index merrily. So far so good.

Year 2016

This was an interesting year. MIDAS-R started the year almost fully deployed. Feb and March saw a dip in stocks though. The Modi effect was wearing off. Some longs were exited. And the system went upto 35% in cash.

Brexit

June saw a mini turbulence as Britishers voted in favor of Brexit. Equity markets saw a couple of days of turbulence and recovered lost gains very fast. After that, uptrend resumed. Until Diwali.

Demonetization

On 8th November, at 8 pm, PM Modi came on TV, and declared all Rs. 500 and Rs. 1000 denomination currency notes null and void from midnight onwards. The nation went into panic mode. There was ultimate secrecy about this announcement. Even the cabinet was apparently told about this only an hour or so before he announced it.

The BJP called it a surgical strike on black money. The opposition called it a monumental blunder. With the benefit of hindsight, its tough for even ardent BJP supporters to call “demon”, as it popularly came to be referred to, a success. The first few days there was utter chaos in the country. After a few days as RBI, Finance Ministry and Banks got their act together. But whether the price was worth it or not will only be known in the longer run. Next morning delivered another surprise.

“The Donald” Trump is in the WH

A scenario which was seen as a huge negative for equities globally came true. Donald Trump”ed”, Hillary Clinton to the WH. It came as a shocker to both candidates. Even Trump himself could not believe it. Neither could Hillary and her supporters. The working class was well and truly fed up of false promises made by “status quo” career politicians like Obama and Hillary. They were willing to look beyond all the negative that Trump represented and elect a true outsider to the WH. And to think that Trump chose to run, because of a joke Obama made on him at a Correspondents Dinner in 2012.

Broader markets had been weak since September 2016. From a high of 8968, Nifty corrected to a low of 7916 or approximately 12% in 3 months. MId Cap Index also corrected by approx 15%. These lows were retested in December 16′. Midas-R was hit by this volatile down move. 4 positions were sold in Nov 16′. And from a MTM high of 26% in Sep 2016, it ended the year at 8.1%. MIDAS on the other hand, ended the year at a healthy 14.6%. This is because markets recovered swiftly by December end. However, the market filter used by MIDAS-R, delayed the re-entry. Insurance has a price. MIDAS on the other hand, which was always in, bore the brunt of the sell-off but also benefited from the V-shaped recovery that followed Trump and Demon.

Performance 2016

Portfolio as at end Dec 2016

MIDAS-R for the first time since inception was beaten by its cousin, MIDAS. Both however, still beat the benchmark index even though MIDAS-R saw a higher drawdown and higher volatility for the year than MIDAS.

However, the year 2016, showed how fickle markets can be not in terms of returns, but in terms of characteristics. For years markets feared Grexit. And out of no where came Brexit. Demon was literally a sucker punch. And the election of Trump, was predicted by most market pundits as a worst case scenario. Stocks would crash, gold would rise and chaos would reign. And yet with each worse case scenario, the time to recovery for equity markets kept getting quicker and faster.

After Trump is Elected – Wall Street welcomes Trump with a bang

How would a human fund manager live with so much cognitive dissonance. Media narratives turn on dime. Fund Managers cannot. Their brains and stomachs turn into mush thinking about such events. Imagine living through them. So as Agent Smith says – Don’t send a human to do a Machine’s job.

Q.E.D.

Disclaimer: Nothing in this blog should be construed as investment advice. This is purely for educational purposes only.  Please consult an investment advisor before investing.

(With inputs from Aayushi Shah)