4 ms·

You can't go back and just say "Oh, this period and that period". That is hindsight bias. It is psychologically comforting but provides no guide to the future.

You can't go back and just say "Oh, this period and that period". That is hindsight bias. It is psychologically comforting but provides no guide to the future. It is particularly foolhardy to place the ending point at 1941. If you didn't know, the period directly after this saw high equity returns because the savings of almost everyone was funnelled by the govt into govt bonds. If you look at how companies were being priced, it is clear why value outperformed.You had profitable companies with $10 of cash trading for $5. You have companies buying back stock at $10 whilst earning $10. Is this comparable to today? It is kind of surprising that someone taking the cycle view of history (this is the "hardcore" historian approach) appears to have not looked at any contemporaneous evidence. A theory has been created in hindsight, no evidence from the period has been produced. This is history at its worst, and investment research at its worst. This post should be called: all the things you don't learn in the CFA program.

jdsully 7y ago
Ford is hovering around a 1.0 price/book ratio and has dropped below that as early as 2018. Value is still out there. Then as now there are many reasons why people won’t pay more at the present time for these stocks.
- Hypx 7y ago
  Yes, many value stocks are near the price floor, and have almost no room left to drop further. Of course, some of these companies are that cheap because they are basically the walking dead of companies (e.g. Gamestop). But for the companies that do have a future, now is a good time to invest in them.
- hogFeast 7y ago
  If you are measuring value with price/book, you have no hope. The conceit of the "value factor" is wrong-headed: trying to measure value quantitatively is like trying to measure the aesthetic appeal of art quantitatively. It just makes no sense. Quant screens turn up the most uninvestable garbage. The only people who invest that way, haven't ever looked properly at the companies they are investing in (I use value quant screens for shorts now, the flow of money into quant means they are usually comically overpriced). More to the point though: value is only determined in hindsight. A company trading on 30x earnings can be cheap. The only "value factor" that is closely correlated to returns is 5yr forward earnings. If you think a computer can predict earnings five years forward...great, but they can't. Also, if you aren't managing $10bn+ you shouldn't be looking at Ford. Don't make it difficult.
  - Hypx 7y ago
    > Quant screens turn up the most uninvestable garbage. The markets right now are filled with uninvestable garbage. It's not a value-only thing. Any viable investment strategy is either a broad based passive investment, or one where you'll have to pick your stocks carefully.
  - jdsully 7y ago
    Value investing as Graham and Dodd described it is a very specific strategy. Future earnings are hard to predict or calculate - but factories, land, and cash exist and have real value that is computable. Right now the market is saying one of two things with a 1.0 price/book: 1) Ford's ongoing business is currently worth $0 OR 2) Ford's capital assets are currently worth $0 It's much easier to make a judgement call on those predicates which are based in the present rather than predict the future.
- robocat 7y ago
  I think "Value" is virtually meaningless in modern businesses. In the past some businesses perhaps depended strongly on capital (e.g. machinery plus cash). Most modern businesses are simply not dependent on the value of hard capital goods, they depend on the value of their services. The book value of large businesses often contains so many instruments or contracts marked to market, that the book value is largely dependent on market value in disguise. Examples: TSMC, ARM, a local restaurant, PWC, etc. Perhaps an exception is real estate companies?
  - kgwgk 7y ago
    If TSMC is Taiwan Semiconductor it doesn't seem a good example: they have in their balance sheet $35bn (more than half of total assets) classified as "property, plant and equipment" and their financial assets are just one tenth of that.
    - robocat 7y ago
      Book value. If you tried to sell the property, plant and equipment (without staff) then the market value would be a lot less. The plant is almost a sunken cost (e.g. 10nm plant for Intel is a sunken cost). The current market cap of TSMC is 210B - related to the value of its future expected earnings compared with other companies.
      - kgwgk 7y ago
        Yes, book value. You said “The __book value__ of large businesses often contains so many instruments or contracts marked to market, that the __book value__ is largely dependent on market value in disguise.” I thought TSMC was supposed to be an example of that, but its assets are cash and old-fashioned capital equipment mostly. Even goodwill is minimal.
UncleEntity 7y ago
> It is particularly foolhardy to place the ending point at 1941. Seems a perfectly logical point since after that the gov't took control of the economy to fuel the war effort and dictated what was produced. Also, how may Axis citizens saw high equity returns on their government bonds?
- hogFeast 7y ago
  It isn't a "perfectly logical point" as I have explained. The study looks at technology, not war (as you are suggesting). It is probably the year in the whole of the century that, in hindsight, explains value returns best...is that coincidental choice? I am not sure what your point is about the Axis either...as with the US and the UK, most people had their savings inflated away. To be totally clear, and this isn't 100% related with the topic: that period is a terrible example because it is reproducible. The assumption with all these studies of long-term returns is that past returns are reproducible. They are not. If you were a normal citizen around 1941, you were forced to invest in govt bonds (banks were forced to buy govt bonds, if you had a deposit you owned govt bonds).
justin66 7y ago
> If you look at how companies were being priced, it is clear why value outperformed. You ought to write a paper. Eugene Fama and Kenneth French have always been pretty upfront about not understanding why value outperforms.
sytelus 7y ago
You buy stock so you can sell at some future date to convert it into cash. This means you must estimate future value of the stock. The value investing as it gets touted implies that you buy stock given its current value. This is incompatible to the goal of selling stock in future because current value may decline to any arbitrary value. Ford may sell at 1.0 price/book today but it is likely they may to load up on debt down the line OR the value of their assets has lower future value. So value investing doesn't mean great investing as much as several people wants us to believe. Every stock purchase is necessarily a speculation and there is no escape from that.

jdsully 7y ago

Ford is hovering around a 1.0 price/book ratio and has dropped below that as early as 2018. Value is still out there. Then as now there are many reasons why people won’t pay more at the present time for these stocks.

Hypx 7y ago
Yes, many value stocks are near the price floor, and have almost no room left to drop further. Of course, some of these companies are that cheap because they are basically the walking dead of companies (e.g. Gamestop). But for the companies that do have a future, now is a good time to invest in them.
hogFeast 7y ago
If you are measuring value with price/book, you have no hope. The conceit of the "value factor" is wrong-headed: trying to measure value quantitatively is like trying to measure the aesthetic appeal of art quantitatively. It just makes no sense. Quant screens turn up the most uninvestable garbage. The only people who invest that way, haven't ever looked properly at the companies they are investing in (I use value quant screens for shorts now, the flow of money into quant means they are usually comically overpriced). More to the point though: value is only determined in hindsight. A company trading on 30x earnings can be cheap. The only "value factor" that is closely correlated to returns is 5yr forward earnings. If you think a computer can predict earnings five years forward...great, but they can't. Also, if you aren't managing $10bn+ you shouldn't be looking at Ford. Don't make it difficult.
- Hypx 7y ago
  > Quant screens turn up the most uninvestable garbage. The markets right now are filled with uninvestable garbage. It's not a value-only thing. Any viable investment strategy is either a broad based passive investment, or one where you'll have to pick your stocks carefully.
- jdsully 7y ago
  Value investing as Graham and Dodd described it is a very specific strategy. Future earnings are hard to predict or calculate - but factories, land, and cash exist and have real value that is computable. Right now the market is saying one of two things with a 1.0 price/book: 1) Ford's ongoing business is currently worth $0 OR 2) Ford's capital assets are currently worth $0 It's much easier to make a judgement call on those predicates which are based in the present rather than predict the future.
robocat 7y ago
I think "Value" is virtually meaningless in modern businesses. In the past some businesses perhaps depended strongly on capital (e.g. machinery plus cash). Most modern businesses are simply not dependent on the value of hard capital goods, they depend on the value of their services. The book value of large businesses often contains so many instruments or contracts marked to market, that the book value is largely dependent on market value in disguise. Examples: TSMC, ARM, a local restaurant, PWC, etc. Perhaps an exception is real estate companies?
- kgwgk 7y ago
  If TSMC is Taiwan Semiconductor it doesn't seem a good example: they have in their balance sheet $35bn (more than half of total assets) classified as "property, plant and equipment" and their financial assets are just one tenth of that.
  - robocat 7y ago
    Book value. If you tried to sell the property, plant and equipment (without staff) then the market value would be a lot less. The plant is almost a sunken cost (e.g. 10nm plant for Intel is a sunken cost). The current market cap of TSMC is 210B - related to the value of its future expected earnings compared with other companies.
    - kgwgk 7y ago
      Yes, book value. You said “The __book value__ of large businesses often contains so many instruments or contracts marked to market, that the __book value__ is largely dependent on market value in disguise.” I thought TSMC was supposed to be an example of that, but its assets are cash and old-fashioned capital equipment mostly. Even goodwill is minimal.

UncleEntity 7y ago

> It is particularly foolhardy to place the ending point at 1941. Seems a perfectly logical point since after that the gov't took control of the economy to fuel the war effort and dictated what was produced. Also, how may Axis citizens saw high equity returns on their government bonds?

hogFeast 7y ago
It isn't a "perfectly logical point" as I have explained. The study looks at technology, not war (as you are suggesting). It is probably the year in the whole of the century that, in hindsight, explains value returns best...is that coincidental choice? I am not sure what your point is about the Axis either...as with the US and the UK, most people had their savings inflated away. To be totally clear, and this isn't 100% related with the topic: that period is a terrible example because it is reproducible. The assumption with all these studies of long-term returns is that past returns are reproducible. They are not. If you were a normal citizen around 1941, you were forced to invest in govt bonds (banks were forced to buy govt bonds, if you had a deposit you owned govt bonds).

justin66 7y ago

> If you look at how companies were being priced, it is clear why value outperformed. You ought to write a paper. Eugene Fama and Kenneth French have always been pretty upfront about not understanding why value outperforms.

sytelus 7y ago

You buy stock so you can sell at some future date to convert it into cash. This means you must estimate future value of the stock. The value investing as it gets touted implies that you buy stock given its current value. This is incompatible to the goal of selling stock in future because current value may decline to any arbitrary value. Ford may sell at 1.0 price/book today but it is likely they may to load up on debt down the line OR the value of their assets has lower future value. So value investing doesn't mean great investing as much as several people wants us to believe. Every stock purchase is necessarily a speculation and there is no escape from that.