I think it's possible people are getting carried away with this example that is circulating widely at the moment. I haven't heard any discussion of the multiple Buffett paid here. It was 1 x earnings. 1 x earnings for a company that was growing revenues at 70% and maintaining margins and management he knew personally. There is the lesson and the secret.
Thanks for the article. I remember puzzling over that explanation that Alice gave. Almost seemed like she was being vague on purpose. I am sure you have read this but check out the 2013 Berkshire letter and the section called Some Thoughts About Investing. He gets very explicit on how he thinks about estimating investment returns.
Although Buffett and Munger acknowledge that the value of a company is essentially its discounted future cash flows they have never used a DCF models.
At the annual Berkshire Hathaway meeting of shareholders in 1996, Charlie Munger said that they use a fingers and toes style of valuation. He went on to say that while Buffett refers to discounted cash flows he had never seen him calculate one.
Buffett joked in response that there are some things that you only do in private!
What Munger was saying is that if a company is not obviously good value, then the opportunity is not attractive enough. If the value of a company does not scream at you then drop the investment idea. You shouldn't need complicated spread sheets and methodologies to help distinguish a good investment idea from a bad one.
Thanks, James. I agree that Buffett doesn't use traditional DCF models - or use WACC, as he and Munger never heard an intelligent conversation about cost of capital - but I do contend that he forecasts and does some form of modelling (mostly in his head).
In a 2022 interview, for example, Todd Combs shared that he and Buffett discuss ideas with the following framework:
"How many names in the S&P are going to be 15x earnings in the next 12 months? How many are going to earn more in five years (using a 90% confidence interval), and how many will compound at 7% (using a 50% confidence interval)?”
If you're determining what a company might earn in one year or five years and are handicapping odds of success, you are making assumptions about the future. Even if they're based on historical trends, you're still making an assumption that the past trends will repeat.
Ideally, those assumptions will be so conservative that the value opportunity will be a no-brainer, but they're still a forecast of some type.
Over time, I've also come to appreciate that you don't need complicated spread sheets to find and determine value (indeed it can have the opposite effect), but I think it's prudent to acknowledge that whether we use a three-statement DCF with a 10-year forecast, rules of thumb, historical rates, or personal hurdle rates that they're all some form of forecasting.
If you are a long term compounder, finding companies that will compound over the long-term (such as Coca-Cola, Gillette or American Express) isn't easy. According to empirical evidence, only 4% of companies meet this criteria: https://rockandturner.substack.com/p/the-4-of-companies-worth-holding
But if you are holding a business across decades, forecasting becomes impossible. Who could have foreseen Covid-19, the Ukranian invasion and its consequential impact on inflation and rates, etc.
So with a long-term time horizon, forecasting becomes futile. Instead, I believe Buffett has a key rule - instead of finding companies that are revolutionizing the world by introducing change, he looks for companies that don't require change to occur for them to be successful (such as Coca-Cola, Gillette or American Express).
Looks like the debate here is over the nature of the forecasting. Asking questions if the sales will grow in the next 5 years and the margin will revert back to mean is one type of forecasting. On the other hand, forecasting 13% sales growth and 11% OPM next year is the other extreme. To each one his own. I am more of a first kind of forecaster. Will the business survive? Will business grow in the future and will margin revert to the mean?
Note that 15% margin on 2 million dollar sales is coming from the author and not Buffett himself. May be he meant some ballpark future scenario to check if he has any upside margin in the investment. We will never know. I do not believe this author can ever understand and read the mind like Buffett.
I think it's possible people are getting carried away with this example that is circulating widely at the moment. I haven't heard any discussion of the multiple Buffett paid here. It was 1 x earnings. 1 x earnings for a company that was growing revenues at 70% and maintaining margins and management he knew personally. There is the lesson and the secret.
Thanks for the article. I remember puzzling over that explanation that Alice gave. Almost seemed like she was being vague on purpose. I am sure you have read this but check out the 2013 Berkshire letter and the section called Some Thoughts About Investing. He gets very explicit on how he thinks about estimating investment returns.
Although Buffett and Munger acknowledge that the value of a company is essentially its discounted future cash flows they have never used a DCF models.
At the annual Berkshire Hathaway meeting of shareholders in 1996, Charlie Munger said that they use a fingers and toes style of valuation. He went on to say that while Buffett refers to discounted cash flows he had never seen him calculate one.
Buffett joked in response that there are some things that you only do in private!
What Munger was saying is that if a company is not obviously good value, then the opportunity is not attractive enough. If the value of a company does not scream at you then drop the investment idea. You shouldn't need complicated spread sheets and methodologies to help distinguish a good investment idea from a bad one.
Thanks, James. I agree that Buffett doesn't use traditional DCF models - or use WACC, as he and Munger never heard an intelligent conversation about cost of capital - but I do contend that he forecasts and does some form of modelling (mostly in his head).
In a 2022 interview, for example, Todd Combs shared that he and Buffett discuss ideas with the following framework:
"How many names in the S&P are going to be 15x earnings in the next 12 months? How many are going to earn more in five years (using a 90% confidence interval), and how many will compound at 7% (using a 50% confidence interval)?”
If you're determining what a company might earn in one year or five years and are handicapping odds of success, you are making assumptions about the future. Even if they're based on historical trends, you're still making an assumption that the past trends will repeat.
Ideally, those assumptions will be so conservative that the value opportunity will be a no-brainer, but they're still a forecast of some type.
Over time, I've also come to appreciate that you don't need complicated spread sheets to find and determine value (indeed it can have the opposite effect), but I think it's prudent to acknowledge that whether we use a three-statement DCF with a 10-year forecast, rules of thumb, historical rates, or personal hurdle rates that they're all some form of forecasting.
If you are a long term compounder, finding companies that will compound over the long-term (such as Coca-Cola, Gillette or American Express) isn't easy. According to empirical evidence, only 4% of companies meet this criteria: https://rockandturner.substack.com/p/the-4-of-companies-worth-holding
But if you are holding a business across decades, forecasting becomes impossible. Who could have foreseen Covid-19, the Ukranian invasion and its consequential impact on inflation and rates, etc.
So with a long-term time horizon, forecasting becomes futile. Instead, I believe Buffett has a key rule - instead of finding companies that are revolutionizing the world by introducing change, he looks for companies that don't require change to occur for them to be successful (such as Coca-Cola, Gillette or American Express).
Looks like the debate here is over the nature of the forecasting. Asking questions if the sales will grow in the next 5 years and the margin will revert back to mean is one type of forecasting. On the other hand, forecasting 13% sales growth and 11% OPM next year is the other extreme. To each one his own. I am more of a first kind of forecaster. Will the business survive? Will business grow in the future and will margin revert to the mean?
Note that 15% margin on 2 million dollar sales is coming from the author and not Buffett himself. May be he meant some ballpark future scenario to check if he has any upside margin in the investment. We will never know. I do not believe this author can ever understand and read the mind like Buffett.
He might have wanted to use Excel, only if..)