The intrinsic value of that bet is $50,000. Making that bet while risking $1 is both extremely risk averse, as well as extremely rewarding. However, risking $49,999 would not be.
Paradox of value investing
- Thread starter Cutten
- Start date
The intrinsic value of that bet is $50,000. Making that bet while risking $1 is both extremely risk averse, as well as extremely rewarding. However, risking $49,999 would not be.
Yeah I agree with most of this. I think it's pretty rational for them to forecast that price will eventually reach fair value. In any case, they can get paid superior returns while they wait (dividends etc). The main question was why buy at $40, not $79.99. Your "real world" considerations explain why (obviously I knew that), but doesn't that just lower the rational bid price? You still have a pretty small bid price to sale price spread - more than 1-2 cents, but certainly not $40 per share.
Would you say then that the "real world" version of value investing would simply be the trading game version, but with the buy-price/sell-price spread widened enough such that it compensated for the risk, uncertainty, and transactions costs involved? I don't see how the spread between buy price (however low that is) and sell price can be very high, because if it was then the quesiton is why not buy at a higher price? Being flat at $41 and long at $79 makes no sense.
It appears to me that the entry price is basically irrelevant, as long as the entry price plus buy price/sell price spread is expected to fall at or below fair value. It would in fact be irrational to demand any discount beyond that on entry, because you would be passing up a certain profit, in return for the *uncertain* possibility that you get an even better deal. It would be like failing to buy a dollar for 99 cents, just because you hope you can get it for 50 cents. That can be rational *only if* you have some credible forecast that the 50 cent offer is likely. In the stock world, uncertainty means maybe you can only offer 50 cents. But if so, then the same uncertainty prevents you rationally holding until 99/100 cents. You gotta sell out lower - maybe 55, maybe 60, maybe even 90 - but certainly not 99/100.
This appears to blow the whole "margin of safety" concept out of the water. Waiting for anything more than a slight bargain appears to be an entirely unjustified (under value investing principles anyway) gamble that even better bargains will appear.
My feeling is that the really demanding bargain hunters are unintentionally engaging in pure speculation on future price. They are gambling that large bargains will appear. This is a pure price prediction with nothing to do with instrinsic value, business principles, or security analysis.
Doesn't their forecast of convergence compel them to buy at $79.99?
Even if real-world assumptions mean a far lower first bid is justified (due to risk, costs etc), don't those same assumptions also compel a sale way below intrinsic value? If stocks are so risky that you need a 50% discount, then the moment that discount narrows to say 45%, 40%, 35%, suddenly you are not getting compensated sufficiently to own it. You required 50% before to compensate for the risk - what changed to make 35% suddenly acceptable?
So I don't think even this forecast of convergence avoids the paradox. Because if their forecast is accurate enough to justifying them holding at $78, $79, $79.99, then it's accurate enough for them to buy in the first place at those same prices. And if the forecast is not accurate, then they can only buy at the margin of safety discount price (e.g. $40). But in this latter case, they must then sell a small amount higher, since the margin of safety will then become too small to justify the risk.
Paradox still intact?
IMO the only way to rationally buy at $40 and sell at $80 is to make two price forecasts. Firstly, a forecast that the bargain price will be reached - without this forecast, you must buy at $79.99 and lower (or $79.99 minus the risk premium). Secondly, a forecast the fair value will be reached - without this, you must sell a few ticks or dollars above your entry price.
These two forecasts are necessary to act in classic value investing style. Yet much teaching in value investing, and many value investors themselves, insist that they do not forecast price. In fact, most think price forecasting is impossible.
Does this mean the whole Graham & Dodd approach is completely irrational and unjustified? And that all investing involves pure price speculation as a core component?
The margin of safety should be enough to compensate the investor for the potential risk that a stock will never reach their assessed intrinsic value. In that case, the stock isn't hold forever but is sold off because eventually some other deal comes along.
I don't consider it to be a forcast when I say that sometime, somewhere a stock will come along that meets my criteria. It's just the way markets work.
I don't consider it to be a forcast when I say that sometime, somewhere a stock will come along that meets my criteria. It's just the way markets work.
OK, now let's back-up a minute. The answers were based on the set-up supplied. Based on the variables and assumptions you supplied, I answered how a "value" investor may look at it.
Now I am not sure if someone would label me a "value" investor or not but my objective is to purchase stock that I believe is currently selling for less than its "value" which doesn't necessarily have to be intrinsic or book value. It also doesn't necessarily mean that I buy low and sell high, it may also mean that I buy high and sell higher.
The gist of what I do is as follows:
I identify stock that I believe will perform better than the averages using my valuation criteria. I don't set the price, the market does that. I sell regardless of gain or loss when my criteria is violated. Performing better than the averages means that the stock I purchase should rise faster than the averages, stay still when the averages decline slightly and decline less than the averages decline. For every 3 stocks that I identify, 2 of them perform better than the averages while the third performs the same or even worse than the averages. By doing this, my portfolio has bested the averages. If I couldn't best the averages, then I would be better off buying an ETF and be done with it.
Joe.
Now I am not sure if someone would label me a "value" investor or not but my objective is to purchase stock that I believe is currently selling for less than its "value" which doesn't necessarily have to be intrinsic or book value. It also doesn't necessarily mean that I buy low and sell high, it may also mean that I buy high and sell higher.
The gist of what I do is as follows:
I identify stock that I believe will perform better than the averages using my valuation criteria. I don't set the price, the market does that. I sell regardless of gain or loss when my criteria is violated. Performing better than the averages means that the stock I purchase should rise faster than the averages, stay still when the averages decline slightly and decline less than the averages decline. For every 3 stocks that I identify, 2 of them perform better than the averages while the third performs the same or even worse than the averages. By doing this, my portfolio has bested the averages. If I couldn't best the averages, then I would be better off buying an ETF and be done with it.
Joe.
Having played Devil's Advocate, I switch sides. I think value investing can be defended on one assumption:
The assumption is that at any given time, there will be some stocks which trade at an identifiable discount of 50% or greater from fair value. Or if there are none now, there will be within a reasonable period of time.
Without this assumption, you can't rationally pass up a stock that is slightly cheap. A slightly cheap stock will have a risk-adjusted return higher than t-bills, or the S&P, or any other efficiently priced asset. The ONLY rational way to not buy it as soon as it falls 1 cent below fair value, is if you expect that it will eventually trade to a much bigger discount.
Note that no value investors wait for a 99.99% discount to fair value. That's because stocks never get THAT cheap. The 50% Graham & Dodd discount is probably a pragmatic level - some stocks in any given year generally do fall to at least half fair value. Thus, the main reason not to buy at a 1% discount is simply that you can find far better opportunities in most years.
There should therefore exist a "typical" max discount, or typical discount range, for any given year. A good value investor should become familiar with the typical maximum discount for say the 6-12 cheapest value stocks per year. That would enable a discount forecast based on past history and the current market environment.
You can see the problem of being *too* demanding by looking at permabears. They will only buy the S&p when it trades at a P/E of 5 and yields 10%+. By being such demanding value investors, they have been in cash since 1933, and vastly underperformed. On the flip side, people who own stocks regardless will frequently own at way over fair value, and hence suffer major bear markets too. There should be a happy medium somewhere between the two.
The assumption is that at any given time, there will be some stocks which trade at an identifiable discount of 50% or greater from fair value. Or if there are none now, there will be within a reasonable period of time.
Without this assumption, you can't rationally pass up a stock that is slightly cheap. A slightly cheap stock will have a risk-adjusted return higher than t-bills, or the S&P, or any other efficiently priced asset. The ONLY rational way to not buy it as soon as it falls 1 cent below fair value, is if you expect that it will eventually trade to a much bigger discount.
Note that no value investors wait for a 99.99% discount to fair value. That's because stocks never get THAT cheap. The 50% Graham & Dodd discount is probably a pragmatic level - some stocks in any given year generally do fall to at least half fair value. Thus, the main reason not to buy at a 1% discount is simply that you can find far better opportunities in most years.
There should therefore exist a "typical" max discount, or typical discount range, for any given year. A good value investor should become familiar with the typical maximum discount for say the 6-12 cheapest value stocks per year. That would enable a discount forecast based on past history and the current market environment.
You can see the problem of being *too* demanding by looking at permabears. They will only buy the S&p when it trades at a P/E of 5 and yields 10%+. By being such demanding value investors, they have been in cash since 1933, and vastly underperformed. On the flip side, people who own stocks regardless will frequently own at way over fair value, and hence suffer major bear markets too. There should be a happy medium somewhere between the two.
Yeah, this is getting to the crux of it. I do consider that to be a forecast of sorts. For example, if you wanted a 99.9% discount, you'd be in cash forever because it never happens. And with no forecast at all, you'd have to buy at 1 cent below fair value.
I think a reasonable forecast is necessary, and it requires empirical justification via experience/back-testing. You can't just pluck a number out of the air and say "I will wait for a 99.9% discount before investing". You need to be realistic and look at past history "I see that each year, at least 20 stocks sold at 50% or lower than reasonable appraisals of their true intrinsic value".
Value investors may benefit significantly if they did a bit more research on how large this valuation discount tends to be. Many value investors seem to think that they are not forecasting the market's behaviour in any way. Yet the scope of their opportunities is critically dependent upon how irrationally bearish markets get. It would seem wise to spend more time looking at how wide discounts get, in what kind of market environment, so that we strike the right balance between being demanding enough to maximise returns and minimise risk, but not so demanding that we stay in cash for 20 years straight.
You nailed it. Personally, I compose a universe of stocks that I think are undervalued and select those that I think have the best ratio of risk to reward. I adapt my criteria to ensure that my universe remains a certain size and make sure that the stocks I pick all possess a catalyst. Ensuring that my universe for potential investments remains a constant size is as important as keeping a margin of safety or determining intrinsic value.
If stocks are so risky that you need a 50% discount, then the moment that discount narrows to say 45%, 40%, 35%, suddenly you are not getting compensated sufficiently to own it. You required 50% before to compensate for the risk - what changed to make 35% suddenly acceptable?
--------------------------------------
The 50% margin may offer a better chance for profit than loss. As the price rises the % of margin of safety decreases which could be considered a lower probability of loss.
Your compensation is less risk.
--------------------------------------
The 50% margin may offer a better chance for profit than loss. As the price rises the % of margin of safety decreases which could be considered a lower probability of loss.
Your compensation is less risk.
No you interpreted my statements incorrectly.
I said I thought the stock was a value at 40. So when it trades I 40 I'm willing to BUY it.
At 41, I'm not willing to buy it. But I am willing to HOLD it if my cost is 40. If the stock trades at 41, my cost is still 40. Nothing changes the fact that what I paid for the stock still remains a value.
On the other hand, if it trades at 41, and you do not already own the stock, it is no longer a buy.
If it trades at 41, that doesn't mean it is a sell.
Put another way, there is a price that is a value...in this case that is 40. There is another price that I would determine as overvalued....and that is not 41.
OldTrader