When Vipul Mathur buys HUL stocks, you buy it too

In a world where there is tremendous fan following for stock pickers and investment strategists, there just aren't enough followers of corporate insiders who are pretty good at timing stock purchases. We develop an algo-trading strategy to exploit the hypothesis that positive alpha can be generated 3-, 6-, and 12- months post large buy transactions by corporate insiders by piggy backing on them.

Vipul Mathur is the VP at Hindustan Unilever Limited in the Fabric Care division. He has been working for HUL for about 19 years now. An IIT-M & IIM-C graduate, he went straight from B-School to HUL and joined the company as a trainee. And that was 19 years ago. HUL is the only company he has ever worked for and probably knows about far more than any other investment guru/equity analyst/investment strategist. When a man like him buys stocks in HUL, you buy it too. No questions asked. Vipul bought Rs.50 Lakhs worth of HUL shares in an open market purchase 3 weeks back.

The pervasiveness of insider trading

Bloomberg carried an article this week on the pervasiveness of insider trading. The report captures a sense that markets in the real world are biased in favour of the corporate elite class. For instance, Sneha Patel, the CEO at Greenwich Lifesciences Inc., made five purchases of her company’s stocks and has earned an average return of 488% on them. Four of those trades preceded the announcement of promising results from a cancer drug trial, the details of which were apparently published on their website beforehand.

Their report also reveals a large fan-following of such successful corporate insider stock traders in the US. In India however, a lot of fan-following is restricted to people like Dolly Khanna or Rakesh Jhunjhunwala. It is rather unclear to me why there isn’t a fan following for Vikram Somany (who bought stocks in Cera Sanitaryware, where he is the MD and Exec Chairman, on 13 occasions and earned an average excess return of 112% over the BSE midcap index after 1 year) or V P Nandakumar (who bought stocks in Manappuram Finance, where he is the CEO, on 25 occasions and made positive excess returns over the midcap index on 16 occasions. He earned an average excess of 92% over the index, 1 year after purchase). My premise is not that these trades are motivated by their access to non-public material information, which would be illegal. But it is that these executives know much more about their businesses than anybody else. And if they are ready and willing to fork out money from their personal savings to buy a meaningful quantity of stocks in the open market, it is a very clear message about the strength of the business. The phrase ‘a big meaningful quantity’ is quite important because some just buy small token amounts to signal their confidence to the market. And you shouldn’t take their word for it. A big meaningful purchase on the other hand makes the message crystal clear. And on those occasions, you should take their lead.

Taking cues from insider trading

At EveryFin, we are in the process of developing a quantitative heuristic-based trading algorithm. And one of the signals that our trading model relies on is the vast repository of insider trading information from BSE/NSE. We did an analysis recently to show you how significant these returns are over 3-, 6-, and 12-months periods.

First, we collected all insider trading information for the top 500 stocks listed in BSE over the last 15 years. Then, we separated the insider purchases into 4 quartiles for each firm. For instance, if promoters have made 4 purchases of 100, 75, 50, and 25 stocks respectively of a particular stock, these trades will get categorised as Top, Lower Top, Uppter Bottom, and Bottom quartiles. We tracked the returns post their purchase date over the next 3-, 6-, and 12-months period to see if the excess returns are statistically significant. We also checked how different the returns are across the four quartiles. The results are below.

Here is a chart showing the average excess return over the midcap index following purchases by insiders.

Stunningly, ALL excess return figures over ALL quartiles over ALL time periods are positive. And at 90% confidence levels, they are more than 0%! That is evidence enough, statistically speaking to believe in the hypothesis that trades following insider trades are expected to deliver positive excess returns. Average excess returns were a full 30% 12 months after purchase in the top quartile of all purchases – in other words, the biggest purchases returned the highest excess returns over the first one-year period.

Median excess returns were also positive and significant. Look at this chart below.

Here is a look at the distribution of excess returns over the three time horizons we examined.

All the three charts overlayed here are decidedly non-normal. The skewness and excess kurtosis factors for the 12-month excess return are 5 and 49 respectively! They are positively skewed with a very long tail on the right. And that is good! This box plot below shows the same trend.

Negative returns are of course bounded by -100% at the bottom. But look at the number of excess returns that are far about the 3X standard deviation mark! Many of the purchases by corporate insiders have extremely handsomely rewarded them over 6 and 12 months past purchase date. The highest goes up to 8X in 12 months!

The probability of making a positive excess return over the index 12 months after a corporate insider makes a big buy transaction (falling in the top quartile) is a stunning 71%!

Making a buck or two - Piggy backing corporate insider traders

How do we use this information to make a buck or two? Well, we can evaluate all insider trading information and buy as soon as we observe a big trade. The expected excess returns over the next 12 months are close to 30%.

But which companies do we go after? All the 500? That’d be a lot of work. It can be without the right technological intervention and that is what we have fixed at EveryFin. Our automated trading system checks for insider transactions and runs this analysis regularly to suggest possible trades. For instance, our current (September 2021 Vintage) insider trading portfolio recommends the following stocks. The company names except HUL have been crossed-out to resist your temptations 😉 We are still in early days.

There is more – some insiders are more successful at trading their companies’ stocks than others are. For instance, The Yamuna Syndicate, Prakash M Patil, Summit Securities Ltd., and Comprehensive Investment & Finance Company Pvt Ltd are examples of promoters who have an excellent track record of timing their stock purchases. And our trading model knows that when one of these insiders buys their stocks, we should sit down and listen to them far more than some others. As a matter of fact, our system has identified 38 corporate insiders like these who it tracks regularly. Our belief is that a stock portfolio built on the foundations of this principle, adequately weighted by the investor’s prior record, is bound to produce positive alpha. Our backtest results are positive and the model is currently live trading. We will post results in a few months’ time! Till then, stay tuned!

Other articles on algo-trading you may be interested in:

Using artificial intelligence in portfolio allocation

A simple application of statistical arbitrage – Mean reversion strategy

Disclosure: I hold stock positions in some of the companies discussed in this post and hence my views may be biased. No part of this article should be considered investment advice. Please do your own due diligence or consult your financial advisor before making an investment.

Share this post

Share on facebook
Share on twitter
Share on linkedin
Share on pinterest
Share on email