The first article of Dr. John Price’s series on what is value in the stock market and how do you measure it. He will look at such methods as Tobin's q-theory, Graham's net asset value, DCF methods, PE methods, technical analysis, etc.
Cynics say that everything has its price: when we are willing to pay enough, we can buy anything. Oscar Wilde said this succinctly in his 1892 play Lady Windermere’s Fan: “What is a cynic? A man who knows the price of everything and the value of nothing.”
Of course, in the stock market everything really does have its price. Whenever a market is open, then there is a clear price at which every stock can be bought or sold.
When it comes to value, as Wilde implies, life becomes much more complicated. For example, it is frequently said that price is what you pay, value is what you get.
The more limited statement, “every stock has a value” is still problematical and contentious. Even consensus on a statement such as “this stock is highly undervalued” could, at best, only be fleeting. Those who agreed with such a statement would keep buying, forcing up the price, until it was no longer “highly undervalued.”
Even if we agree that every stock has a value, there would likely be much less agreement over just what this value is. To see this, suppose for the moment we all agreed on the value of a stock. Let us suppose that it is $20. Then buyers would only bid at $20 or less while sellers would only offer it at $20 or more. Hence the only trades would be at the “value” price of $20 so there would be little point in any buying or selling. The business of the stock market would come to a halt.
You can actually think of the stock market as buying and selling value. If you are of the opinion that the price of a stock represents sufficiently good value, then you may decide to buy. Of course, the person on the other side thinks that the cash from such a transaction represents better value than owning the stock, and so we have a seller.
The curious thing is that both sides see the other person as giving up value and passing it over to them. The nub of the problem is that everyone has their own idea of what represents value.
In this series of articles we are going to look at a range of interlocking questions regarding value in the stock market and how to calculate it. A preliminary list of these questions is:
Does every stock have a value?
What is the value of a stock?
What methods are used to calculate value?
What purpose does calculating value serve?
By looking at questions such as these, the ultimate aim of this series is to provide a framework to answer the fundamental question whether, at any given time for any given stock, it makes sense to buy, to sell, or to do nothing.
The whole question of value is extremely complicated. If we are not careful, we could sink in a quagmire of opinions, analyses, claims and counter-claims about the meaning of value.
To keep things within manageable limits, mostly I will only consider the calculation or estimation of value in monetary terms. There will be times, however, when we will wander into the areas where, for social, ethical and personal reasons, investors may be willing to give up some monetary value for other forms of value.
For example, regularly Warren Buffett, the Chairman and CEO of Berkshire Hathaway, writes in the company’s annual report that Charlie Munger, the Vice Chairman, and he have an “ attitude … that hurts our financial performance: Regardless of price, we have no interest at all in selling any good businesses that Berkshire owns. We are also very reluctant to sell sub-par businesses as long as we expect them to generate at least some cash and as long as we feel good about their managers and labor relations.”
This is a clear statement that Buffett and Munger give value to continuity of holdings even if it takes away from monetary value.
There are two very large obstacles that everyone has to deal with to come to grips with the problem of value: Firstly, the large number of methods for calculating value with their various assumptions and inputs and, secondly, behavioral biases.
Valuation methods
Regarding the first obstacle, there are dozens of methods all claiming to find the “value” of stock and to answer the question of whether to buy, to sell, or to do nothing. The methods have been developed by everyone from enthusiastic amateurs to Nobel laureates. This might not be so bad if they all, at least approximately, agreed with each other. If only life was so simple!
Fortunately the methods fall into a number of major classes and so what I will do is look at key examples in these classes.
There are two things to note right at the outset. The first is that to see whether a purchase actually represents good value requires the passing of time. No matter what sort of bargain you think it is, it is only after a certain amount of time that you can be certain of this. Andrew Smithers of Smithers & Co refers to this as ‘hindsight value’.
Suppose you purchase a used car for a price that you consider is an absolute ‘steal’. It is only after owning it, and most likely using it, for a period of time that you can know whether this is true.
In rare circumstances the period of time may only be minutes. For example, as soon as you have signed the papers, you are offered a higher price which you take. This would be akin to the price of a stock rising so significantly as soon as you have made your purchase that you decide to sell it straight away.
More likely you will own the car for a number of years after which you will resell it or scrap it. When you have done this, then you can decide whether your original purchase provided value or not. You would do this taking into account the service the car has provided, the running costs and the sale price (which would be zero in the case of scrapping the car).
This is similar to purchasing a stock and holding it for a period of time. When you sell, taking into account the dividends that you have received and the final selling price, you can say whether or not the original purchase represented value.
There is a third scenario where you know whether or not you made a good choice before selling or scrapping the car. It is the case where the service that you received from the car grows to a level greater than what you originally paid for it. In this case you know that you have received value even if you neither sell the car nor scrap it.
In the stock market this would be similar to the case where the dividends that you have received are such that they have covered the original purchase price.
The point is that no matter how confident you are at the time of the transaction, value is always uncertain. It is only with hindsight that that this uncertainty can be reduced or even eliminated.
The second thing to note is that each method for calculating or estimating value has its own set of input variables. In a few cases these inputs can be checked immediately such as the current debt of a business or its level of earnings.
In other cases they can only be verified with hindsight. Examples could be the growth rate of earnings over the next five years. It is only after five years that we can verify the accuracy of such an input.
There is a third set of inputs that makes parents who establish trusts and holding companies for their children and grandchildren look like they are suffering short-termism. These inputs are impossible to verify. This bizarre situation applies to the so-called stable or infinite growth phase in dividend discount models or discount cash flow models where inputs require the growth of variables such as dividends ‘out to infinity’. Clearly no time period, whether it be months, years or decades, will be sufficient in length to verify such inputs.
Behavioral biases
The second major obstacle to calculating value is that extensive research shows that we all have unconscious biases. This means we have to face problems such as:
We may have different biases about the most suitable method to calculate value.
Even if there is agreement on the method, we may have different biases on the values of the inputs.
Even if there is agreement on the methods and the inputs, we may have different biases on how to interpret the results and what course of action should be taken.
Biases in the market place come under the general title of behavioral investing. They can be grouped under the areas of (1) heuristic simplification covering the ways we introduce various rules of thumb in decision making, (2) frame dependence covering the ways in which decisions depend on the form or setting of the problem, and (3) social biases covering the ways we are influenced by others.
In a later article in this series I will examine these areas in more detail being content for now just to give a single example. In their key 1974 paper Amos Tversky and Nobel laureate Daniel Kahneman described an experiment where people were asked questions such as the percentage of African nations in the United Nations. Before they answered a wheel of fortune with the numbers 1 to 100 was spun in front of them. The wheel was rigged to give only the results 10 or 65 on each spin.
The median response to the preceding question for those who saw the wheel display the number 10 was 25, and the median response for those that saw 65 was 45. Even though the participants knew that the data was irrelevant and random, it still influenced their answers. This effect is known as anchoring.
In the context of valuation models, investors may anchor their estimates on input variables using simplified or irrelevant information. For example, we may see that the price-to-earnings ratio is high for a number of stocks and conclude that it should be high for the stock under consideration. Anchoring is an example of heuristic simplification.
Time, Chance and Size
Even when we restrict ourselves to value in a financial setting and overcome — or, at least recognize — our biases, there are still three difficulties: time, chance and size. Any final assignment of value needs to incorporate and make allowance for these three attributes.
Consider time. Above we talked about the importance of time in being able to make a final evaluation whether a strategy provided value or not. Time also enters when we have to compare events that occur in different periods.
Suppose we are offered $100 now or a guaranteed $100 in one year. The rational choice would be to take the immediate payment. The reason is that the $100 could be invested in a guaranteed bank deposit in order to receive more than $100 in one year.
Suppose instead we are offered $100 now or a guaranteed $200 in one year. The rational choice would be to take the $200 in one year since it would be extremely unlikely that an immediate $100 could be converted to $200 in one year. This means that somewhere between $100 and $200 there is an equilibrium amount. For this amount we see it as equal value to accept the $100 now or the equilibrium amount in one year. In technical terms, what we are talking about is the time-value of money and how it has to be incorporated in any value calculations or judgments.
Next, suppose that the payment in one year is not guaranteed. For example, suppose that there is a 90% chance of being paid in full and a 10% chance of a complete default. We would increase the preceding equal-value equilibrium amount to compensate for the chance or probability of not receiving anything. Of course, in real life situations it is impossible to put such precise numbers on probabilities. The books Fooled by Randomness and The Black Swan by Nassim Nicholas Taleb make the unanticipated effects of randomness very clear.
To see how size enters, suppose you are offered a choice with equal chance between receiving $5 and paying $1. Most people (in a developed country) would accept this choice reasoning that losing $1 is not a major consequence especially against the chance of gaining $5. Suppose the amounts were $500 and $100. Then fewer people would accept the choice since losing $100 is a more serious outcome than losing $10. As the amounts increase, say to $5,000 and $1,000 or $50,000 and $10,000, the number of people accepting this choice would decline.
In summary, even if we are completely happy with both the method we intend to use to assign value to an action and the input variables, what value we give it also depends on the time of the action, the probabilities involved, and the size of the transaction.
____
Having looked at some of the issues regarding value and stated our goal to provide a framework to answer the fundamental question whether, at any given time for any given stock, it makes sense to buy, to sell, or to do nothing, in the next article in this series I will look at value as measured through the work of the Nobel laureate James Tobin and his q-theory. One of the surprising consequences of this approach is that value in the stock market does not depend on short-term and long-term interest rates.
Cynics say that everything has its price: when we are willing to pay enough, we can buy anything. Oscar Wilde said this succinctly in his 1892 play Lady Windermere’s Fan: “What is a cynic? A man who knows the price of everything and the value of nothing.”
Of course, in the stock market everything really does have its price. Whenever a market is open, then there is a clear price at which every stock can be bought or sold.
When it comes to value, as Wilde implies, life becomes much more complicated. For example, it is frequently said that price is what you pay, value is what you get.
The more limited statement, “every stock has a value” is still problematical and contentious. Even consensus on a statement such as “this stock is highly undervalued” could, at best, only be fleeting. Those who agreed with such a statement would keep buying, forcing up the price, until it was no longer “highly undervalued.”
Even if we agree that every stock has a value, there would likely be much less agreement over just what this value is. To see this, suppose for the moment we all agreed on the value of a stock. Let us suppose that it is $20. Then buyers would only bid at $20 or less while sellers would only offer it at $20 or more. Hence the only trades would be at the “value” price of $20 so there would be little point in any buying or selling. The business of the stock market would come to a halt.
You can actually think of the stock market as buying and selling value. If you are of the opinion that the price of a stock represents sufficiently good value, then you may decide to buy. Of course, the person on the other side thinks that the cash from such a transaction represents better value than owning the stock, and so we have a seller.
The curious thing is that both sides see the other person as giving up value and passing it over to them. The nub of the problem is that everyone has their own idea of what represents value.
In this series of articles we are going to look at a range of interlocking questions regarding value in the stock market and how to calculate it. A preliminary list of these questions is:
Does every stock have a value?
What is the value of a stock?
What methods are used to calculate value?
What purpose does calculating value serve?
By looking at questions such as these, the ultimate aim of this series is to provide a framework to answer the fundamental question whether, at any given time for any given stock, it makes sense to buy, to sell, or to do nothing.
The whole question of value is extremely complicated. If we are not careful, we could sink in a quagmire of opinions, analyses, claims and counter-claims about the meaning of value.
To keep things within manageable limits, mostly I will only consider the calculation or estimation of value in monetary terms. There will be times, however, when we will wander into the areas where, for social, ethical and personal reasons, investors may be willing to give up some monetary value for other forms of value.
For example, regularly Warren Buffett, the Chairman and CEO of Berkshire Hathaway, writes in the company’s annual report that Charlie Munger, the Vice Chairman, and he have an “ attitude … that hurts our financial performance: Regardless of price, we have no interest at all in selling any good businesses that Berkshire owns. We are also very reluctant to sell sub-par businesses as long as we expect them to generate at least some cash and as long as we feel good about their managers and labor relations.”
This is a clear statement that Buffett and Munger give value to continuity of holdings even if it takes away from monetary value.
There are two very large obstacles that everyone has to deal with to come to grips with the problem of value: Firstly, the large number of methods for calculating value with their various assumptions and inputs and, secondly, behavioral biases.
Valuation methods
Regarding the first obstacle, there are dozens of methods all claiming to find the “value” of stock and to answer the question of whether to buy, to sell, or to do nothing. The methods have been developed by everyone from enthusiastic amateurs to Nobel laureates. This might not be so bad if they all, at least approximately, agreed with each other. If only life was so simple!
Fortunately the methods fall into a number of major classes and so what I will do is look at key examples in these classes.
There are two things to note right at the outset. The first is that to see whether a purchase actually represents good value requires the passing of time. No matter what sort of bargain you think it is, it is only after a certain amount of time that you can be certain of this. Andrew Smithers of Smithers & Co refers to this as ‘hindsight value’.
Suppose you purchase a used car for a price that you consider is an absolute ‘steal’. It is only after owning it, and most likely using it, for a period of time that you can know whether this is true.
In rare circumstances the period of time may only be minutes. For example, as soon as you have signed the papers, you are offered a higher price which you take. This would be akin to the price of a stock rising so significantly as soon as you have made your purchase that you decide to sell it straight away.
More likely you will own the car for a number of years after which you will resell it or scrap it. When you have done this, then you can decide whether your original purchase provided value or not. You would do this taking into account the service the car has provided, the running costs and the sale price (which would be zero in the case of scrapping the car).
This is similar to purchasing a stock and holding it for a period of time. When you sell, taking into account the dividends that you have received and the final selling price, you can say whether or not the original purchase represented value.
There is a third scenario where you know whether or not you made a good choice before selling or scrapping the car. It is the case where the service that you received from the car grows to a level greater than what you originally paid for it. In this case you know that you have received value even if you neither sell the car nor scrap it.
In the stock market this would be similar to the case where the dividends that you have received are such that they have covered the original purchase price.
The point is that no matter how confident you are at the time of the transaction, value is always uncertain. It is only with hindsight that that this uncertainty can be reduced or even eliminated.
The second thing to note is that each method for calculating or estimating value has its own set of input variables. In a few cases these inputs can be checked immediately such as the current debt of a business or its level of earnings.
In other cases they can only be verified with hindsight. Examples could be the growth rate of earnings over the next five years. It is only after five years that we can verify the accuracy of such an input.
There is a third set of inputs that makes parents who establish trusts and holding companies for their children and grandchildren look like they are suffering short-termism. These inputs are impossible to verify. This bizarre situation applies to the so-called stable or infinite growth phase in dividend discount models or discount cash flow models where inputs require the growth of variables such as dividends ‘out to infinity’. Clearly no time period, whether it be months, years or decades, will be sufficient in length to verify such inputs.
Behavioral biases
The second major obstacle to calculating value is that extensive research shows that we all have unconscious biases. This means we have to face problems such as:
We may have different biases about the most suitable method to calculate value.
Even if there is agreement on the method, we may have different biases on the values of the inputs.
Even if there is agreement on the methods and the inputs, we may have different biases on how to interpret the results and what course of action should be taken.
Biases in the market place come under the general title of behavioral investing. They can be grouped under the areas of (1) heuristic simplification covering the ways we introduce various rules of thumb in decision making, (2) frame dependence covering the ways in which decisions depend on the form or setting of the problem, and (3) social biases covering the ways we are influenced by others.
In a later article in this series I will examine these areas in more detail being content for now just to give a single example. In their key 1974 paper Amos Tversky and Nobel laureate Daniel Kahneman described an experiment where people were asked questions such as the percentage of African nations in the United Nations. Before they answered a wheel of fortune with the numbers 1 to 100 was spun in front of them. The wheel was rigged to give only the results 10 or 65 on each spin.
The median response to the preceding question for those who saw the wheel display the number 10 was 25, and the median response for those that saw 65 was 45. Even though the participants knew that the data was irrelevant and random, it still influenced their answers. This effect is known as anchoring.
In the context of valuation models, investors may anchor their estimates on input variables using simplified or irrelevant information. For example, we may see that the price-to-earnings ratio is high for a number of stocks and conclude that it should be high for the stock under consideration. Anchoring is an example of heuristic simplification.
Time, Chance and Size
Even when we restrict ourselves to value in a financial setting and overcome — or, at least recognize — our biases, there are still three difficulties: time, chance and size. Any final assignment of value needs to incorporate and make allowance for these three attributes.
Consider time. Above we talked about the importance of time in being able to make a final evaluation whether a strategy provided value or not. Time also enters when we have to compare events that occur in different periods.
Suppose we are offered $100 now or a guaranteed $100 in one year. The rational choice would be to take the immediate payment. The reason is that the $100 could be invested in a guaranteed bank deposit in order to receive more than $100 in one year.
Suppose instead we are offered $100 now or a guaranteed $200 in one year. The rational choice would be to take the $200 in one year since it would be extremely unlikely that an immediate $100 could be converted to $200 in one year. This means that somewhere between $100 and $200 there is an equilibrium amount. For this amount we see it as equal value to accept the $100 now or the equilibrium amount in one year. In technical terms, what we are talking about is the time-value of money and how it has to be incorporated in any value calculations or judgments.
Next, suppose that the payment in one year is not guaranteed. For example, suppose that there is a 90% chance of being paid in full and a 10% chance of a complete default. We would increase the preceding equal-value equilibrium amount to compensate for the chance or probability of not receiving anything. Of course, in real life situations it is impossible to put such precise numbers on probabilities. The books Fooled by Randomness and The Black Swan by Nassim Nicholas Taleb make the unanticipated effects of randomness very clear.
To see how size enters, suppose you are offered a choice with equal chance between receiving $5 and paying $1. Most people (in a developed country) would accept this choice reasoning that losing $1 is not a major consequence especially against the chance of gaining $5. Suppose the amounts were $500 and $100. Then fewer people would accept the choice since losing $100 is a more serious outcome than losing $10. As the amounts increase, say to $5,000 and $1,000 or $50,000 and $10,000, the number of people accepting this choice would decline.
In summary, even if we are completely happy with both the method we intend to use to assign value to an action and the input variables, what value we give it also depends on the time of the action, the probabilities involved, and the size of the transaction.
____
Having looked at some of the issues regarding value and stated our goal to provide a framework to answer the fundamental question whether, at any given time for any given stock, it makes sense to buy, to sell, or to do nothing, in the next article in this series I will look at value as measured through the work of the Nobel laureate James Tobin and his q-theory. One of the surprising consequences of this approach is that value in the stock market does not depend on short-term and long-term interest rates.
