As I’m currently working on a series on the European ecosystem in Macros, it’s really nice to read an article like this, it gives a lot of insight. Thank you!
I may have missed it but did you account for survivorship bias? Do all these companies exist today? There’s probably several that were 100xers but then got bought out or delisted. Right now there may be say 5000 companies that have been trading since 1980, but in that same period there could have been 10000 that went bankrupt or taken off the exchange.
I pulled the list of every stock currently traded at the exchanges in the available countries from the data vendor. Then downloaded the OHLC data from every stock currently traded.
Thus all stocks are actively traded right now.
As for survivorship bias, after collecting and cleaning all data I noticed that the data vendor provides delisted stocks too. I decided for now not to redo cleaning by hand, thus delisted stocks are not included. Maybe the list of identified 100 baggers would be longer.
How did you account for the various currencies? The Euro was only introduced in 1999. I suppose the data will just have divided the old national currency prices by the fixed exchange rate from Dec1998. But currencies did fluctuate in the 20 years before that. Maybe using a rather stable currency like DM and converting stock prices pre98 from the national currency into DM and then into € at the dec98 rate could be a workaround. Not perfect of course, but probably more realistic. Also the € was only introduced later in some countries (Greece, Baltics...).
In my experience, often datasets miss special dividends, which can have a big effect on 'adjusted' prices.
If you have the stock price data of LVMH and the predecessors LV group and Moet&Chandon starting from 1980, I would be very interested! 🙏
Thank you for your thoughtful reply. The stock data is adjusted for currency changes by the exchanges. My data provider got the data from original source like CBOE and Nasdaq.
Most stock data pre98 came from England, France, Germany, the Netherlands and the Nordic countries.
To be honest, the adjusted had some other inconsistencies around the split dates. I also tried with an algorithm to adjust the price myself for splits only. The inconsistencies were too much. Old split dates were also had to verify. Especially, the companies with ADR representation in America like ASML.
Finding data cheap and *very* reliable is hard.
If I would do the study again, then I want the raw data directly from the exchanges self.
I am not around my laptop this weekend, but set a reminder to find the LVMH data. I am afraid that the dataset hasn’t the predecessors separately. It only has the current listed stocks. Also another ‘flaw’ in the study. Merged companies have a longer runway to 100-fold, but aren’t reflected in the data.
From my experience it is mostly impossible to obtain historical data from predecessors to merged companies, if it is more than 15-20 years in the past. Only chance would maybe be old stock manuals or newspaper clippings where daily prices are reported, but that would really be a huge amount of work.
I would say doing this work at scale for hundreds of companies is an impossible task for an individual. It can be done for a few companies if you make it a labour of love (I have done it for a few).
Congratulation for your scientific work! The method and rhe results should be part of the educational material in universities, and trigger more research.
While I am sitting in a spec sit investment (an EU CRMA designated huge mine project) with a seemingly sharp policy change (total U turn in a country's stance to such projects) waiting for immiment permissoons, and just calculated the chances and levels to reach from last years basea a 50-100 x bagging (astonishingly quite high)... esp with the paralel price blow up in PM prices.. wonder how mining is not included in your list...
Thank you. I didn’t specifically address the mining industry as I am not acquainted with the industry. But in the list of identified 100 baggers, there are some companies in the metals and mining industry.
And as you suggested, it is very likely that share price increase comes from the precious metals price surge.
There are a couple of factors which do not necessarily have to do with the population size:
- Sweden has some world class companies in boring sectors. For example, Mips develops, manufactures and sells helmet safety systems for cyclists to construction engineers. (Not a 100 bagger, but grows fast)
- Investing among the Swedish population is popular due to governmental education in the 90s and 2000s.
- Companies are owner-operated. There is lots of insider buying in Sweden. Thus management makes the decisions based on their and shareholders' interest. Which results in growing per-share value.
- The Swedish market attracts foreign companies to IPO in Sweden. The Financial Times recently reported that this accelerated due to Brexit, because before London was the popular market to do so.
Despite the different currency, it is an interesting market for investors.
I appreciate your scientific approach to this project. It would be interesting to look at mean financial variables/ratios in your 100x cohort at the starting periods, to see if any predictive indights can be determined.
That's a great idea and made a note of your suggestion. Maybe I will expand on the study later on and include statistical data. If there is statistical significance, then I can imagine that it would be useful with screener filters.
As I’m currently working on a series on the European ecosystem in Macros, it’s really nice to read an article like this, it gives a lot of insight. Thank you!
You’re welcome, I look forward to read your series!
I may have missed it but did you account for survivorship bias? Do all these companies exist today? There’s probably several that were 100xers but then got bought out or delisted. Right now there may be say 5000 companies that have been trading since 1980, but in that same period there could have been 10000 that went bankrupt or taken off the exchange.
I pulled the list of every stock currently traded at the exchanges in the available countries from the data vendor. Then downloaded the OHLC data from every stock currently traded.
Thus all stocks are actively traded right now.
As for survivorship bias, after collecting and cleaning all data I noticed that the data vendor provides delisted stocks too. I decided for now not to redo cleaning by hand, thus delisted stocks are not included. Maybe the list of identified 100 baggers would be longer.
I hope it answers your questions
Thanks for the reply! That makes sense, I guess it’d be hard to mod that out!
Can you find common metrics that would be useful in picking future multi baggers such as PEG ratio; RoE etc?
I haven’t yet, but planned to include this in the second part. Collecting the historical data proves to be more difficult.
In the meantime, you can check out the paper “The Alchemy of Multibagger Stocks” by Anna Yartseva. She has proven that some metrics do correlate.
Good job, very interesting project.
How did you account for the various currencies? The Euro was only introduced in 1999. I suppose the data will just have divided the old national currency prices by the fixed exchange rate from Dec1998. But currencies did fluctuate in the 20 years before that. Maybe using a rather stable currency like DM and converting stock prices pre98 from the national currency into DM and then into € at the dec98 rate could be a workaround. Not perfect of course, but probably more realistic. Also the € was only introduced later in some countries (Greece, Baltics...).
In my experience, often datasets miss special dividends, which can have a big effect on 'adjusted' prices.
If you have the stock price data of LVMH and the predecessors LV group and Moet&Chandon starting from 1980, I would be very interested! 🙏
Thank you for your thoughtful reply. The stock data is adjusted for currency changes by the exchanges. My data provider got the data from original source like CBOE and Nasdaq.
Most stock data pre98 came from England, France, Germany, the Netherlands and the Nordic countries.
To be honest, the adjusted had some other inconsistencies around the split dates. I also tried with an algorithm to adjust the price myself for splits only. The inconsistencies were too much. Old split dates were also had to verify. Especially, the companies with ADR representation in America like ASML.
Finding data cheap and *very* reliable is hard.
If I would do the study again, then I want the raw data directly from the exchanges self.
I am not around my laptop this weekend, but set a reminder to find the LVMH data. I am afraid that the dataset hasn’t the predecessors separately. It only has the current listed stocks. Also another ‘flaw’ in the study. Merged companies have a longer runway to 100-fold, but aren’t reflected in the data.
From my experience it is mostly impossible to obtain historical data from predecessors to merged companies, if it is more than 15-20 years in the past. Only chance would maybe be old stock manuals or newspaper clippings where daily prices are reported, but that would really be a huge amount of work.
I would say doing this work at scale for hundreds of companies is an impossible task for an individual. It can be done for a few companies if you make it a labour of love (I have done it for a few).
I subscribed to your newsletter but was not sent the 1 page checklist to find multi-baggers.
Can you please send it to me?
Thank you.
Mark
I sent it to your email
Great work! I can imagine how hard was to find all that data, congrats.
Thank you! It was indeed. Stay tuned for part 2
Thanks for a great study.
I’m glad you enjoyed it
Congratulation for your scientific work! The method and rhe results should be part of the educational material in universities, and trigger more research.
While I am sitting in a spec sit investment (an EU CRMA designated huge mine project) with a seemingly sharp policy change (total U turn in a country's stance to such projects) waiting for immiment permissoons, and just calculated the chances and levels to reach from last years basea a 50-100 x bagging (astonishingly quite high)... esp with the paralel price blow up in PM prices.. wonder how mining is not included in your list...
Thank you. I didn’t specifically address the mining industry as I am not acquainted with the industry. But in the list of identified 100 baggers, there are some companies in the metals and mining industry.
And as you suggested, it is very likely that share price increase comes from the precious metals price surge.
Great work, thanks for sharing!
You’re welcome!
fantastic post, thanks.
Thank you, stay tuned for part 2
Why is Sweden such an outperformer? It's got the second most amount of 100-baggers but has 1/6th the population of Britain. Very interesting.
There are a couple of factors which do not necessarily have to do with the population size:
- Sweden has some world class companies in boring sectors. For example, Mips develops, manufactures and sells helmet safety systems for cyclists to construction engineers. (Not a 100 bagger, but grows fast)
- Investing among the Swedish population is popular due to governmental education in the 90s and 2000s.
- Companies are owner-operated. There is lots of insider buying in Sweden. Thus management makes the decisions based on their and shareholders' interest. Which results in growing per-share value.
- The Swedish market attracts foreign companies to IPO in Sweden. The Financial Times recently reported that this accelerated due to Brexit, because before London was the popular market to do so.
Despite the different currency, it is an interesting market for investors.
I appreciate your scientific approach to this project. It would be interesting to look at mean financial variables/ratios in your 100x cohort at the starting periods, to see if any predictive indights can be determined.
That's a great idea and made a note of your suggestion. Maybe I will expand on the study later on and include statistical data. If there is statistical significance, then I can imagine that it would be useful with screener filters.
Great post!
Interesting study! Looking forward to read part two.
I am glad to hear!
Really enjoyed this article. As a European investor, this kind of work is especially valuable and refreshing to see.