I posted this dashboard here back in March, but with the recent investigations into insider trading by Kodak and the USPS Postmaster General I thought it was very relevant once again.
There are a lot of sites out there which list off every insider transaction. That’s good if you want to find more detail on a specific case of suspected insider trading, but it makes it difficult to quickly identify companies with suspicious activity.
Instead of displaying data at the transaction-level, I decided to aggregate transactions from a 6-month rolling period. I then created the “Insider Confidence Index”, which measures a company’s insider trading activity relative to their total market volume.
I also created a graph showing daily net purchases/sales across all S&P 500 insiders, to give an idea of the aggregate sentiment among insiders. It was interesting to see a huge spike in selling back in early February, and then again a couple of weeks ago.
I’d love to hear any thoughts or suggestions you have on how to improve this dashboard, and if you’d prefer to see the data at a transaction-level.
> There are a lot of sites out there which list off every insider transaction.
Perhaps pedantic, but those sites only list registered or declared insider transactions. "Insider Trading" - the illegal kind, is trading off of non-public information and can be done by non insiders (who wouldn't have SEC-declared transactions).
Correct -- it is quite common, however, to see lists of "Insider trades" which are declared trades from known insiders.
Peter Lynch is famous for stating: "insiders might sell their shares for any number of reasons, but they buy them for only one: they think the price will rise." Keeping an eye on the trade flow from insiders is interesting at a minimum and at least perhaps useful.
Option expiry is only a problem for employees. Senior management, who are the ones to watch because of their access to sensitive information, typically get options refreshed or amended.
Typically you are overweight in the company you founded/have been granted lots of options in. So there are liquidity/risk management reasons to sell even if you aren't pessimistic about share price.
I hate that you are getting downvoted for this (and I will for this). Needing evidence used to be the penultimate expectation of every hacker news comment. Now its witticisms wrought with extreme prejudice. And even you asking to maybe see a study where someone determined that there was indeed a causal relationship between stock prices and the signal the previous commenters claimed is somehow grounds for an attack on your karma.
A 10b5-1 plan can be cancelled, and cancelling one when in possession of inside information is not insider trading.
Could you set up one plan to buy shares and one to sell them, and cancel whichever one isn't favourable? No. There's a clause that requires you to have set up the plan in good faith. But if you planned to sell your stock, and get once-in-a-lifetime news the price is going up tomorrow, it's completely fine to cancel it.
Maybe it's not currently illegal, but shouldn't it be? Once in a lifetime news on a stock price about to go up sure sounds like material non public information to me.
OP just gave an extreme example but if the CEO saw sales was on a promising trend (or other reasons that don’t count as insider trading) he/she might cancel the sell. And it would be useful to see that.
Is there a good reason that a CEO should be able to use private sales data to inform their trades? (in this case, to cancel a previously scheduled trade)
There is no good reason. It's really a loophole in the law. An understandable one, because cancelling a trade is not legally the same as making a trade, even if financially it's the same. And one that's small enough not to have been worth closing. You won't get away with cancelling your plan every time sales data is bad, you should wait for the once-in-several-lifetimes opportunity when you get game-changing news and you were just about to sell a large quantity of stock.
You can come up with a bigger conspiracy theory than this, but the more boring explanation looks true.
Don't use insider trading as a short term signal. Most of these transactions are contractually fixed months in advance exactly to prevent the insiders from actively trading on current knowledge.
i.e. They commit to sell 100 shares at whatever the price will be in a months time. So it does have some predictive power but only on a very long timescale. There is also a lot of noise from other factors. e.g. Maybe the exec needs cash for his kids uni. Or he feels his net worth is too concentrated in one company.
Side question, anyone know the easiest way to lookup how many shares an individual insider currently holds? Do I need to go through the most recent 10k/q or is there somewhere else I should be looking?
38 comments
[ 2.9 ms ] story [ 57.0 ms ] threadThere are a lot of sites out there which list off every insider transaction. That’s good if you want to find more detail on a specific case of suspected insider trading, but it makes it difficult to quickly identify companies with suspicious activity.
Instead of displaying data at the transaction-level, I decided to aggregate transactions from a 6-month rolling period. I then created the “Insider Confidence Index”, which measures a company’s insider trading activity relative to their total market volume.
I also created a graph showing daily net purchases/sales across all S&P 500 insiders, to give an idea of the aggregate sentiment among insiders. It was interesting to see a huge spike in selling back in early February, and then again a couple of weeks ago.
I’d love to hear any thoughts or suggestions you have on how to improve this dashboard, and if you’d prefer to see the data at a transaction-level.
Working on similar things for the data tornado that is crypto.
Any idea if this is planned share selling? Six month gap between the spikes would suggest this methinks.
Perhaps pedantic, but those sites only list registered or declared insider transactions. "Insider Trading" - the illegal kind, is trading off of non-public information and can be done by non insiders (who wouldn't have SEC-declared transactions).
Peter Lynch is famous for stating: "insiders might sell their shares for any number of reasons, but they buy them for only one: they think the price will rise." Keeping an eye on the trade flow from insiders is interesting at a minimum and at least perhaps useful.
Insider buying is much stronger.
Any evidence (anecdotal or data-driven) to support this?
Could you set up one plan to buy shares and one to sell them, and cancel whichever one isn't favourable? No. There's a clause that requires you to have set up the plan in good faith. But if you planned to sell your stock, and get once-in-a-lifetime news the price is going up tomorrow, it's completely fine to cancel it.
You can come up with a bigger conspiracy theory than this, but the more boring explanation looks true.
EDIT: Made some changes to the database querying and caching, site should be much more fast and stable now
i.e. They commit to sell 100 shares at whatever the price will be in a months time. So it does have some predictive power but only on a very long timescale. There is also a lot of noise from other factors. e.g. Maybe the exec needs cash for his kids uni. Or he feels his net worth is too concentrated in one company.