Candy bars and movie tickets

Arnold Kling has posted an entry on the topic of stock options. His entry gives much food for thought, and I will post a thorough reply in the next couple of days.

In the meantime, a question about an underlying issue : Are there any exchanges/giveaways in which the value of one party's gain is radically different the value of the other party's loss, even though the object transferred is the same? A couple of examples that may be analagous to a limited degree:

kitkat.gif• Suppose that for whatever reason I own every chocolate candy bar in Boston. Nobody else owns any candy bars. My apartment is literally sprawling with candy bars. My closet is jampacked with thousands of them. There are candy bars lying around on the coffee table, on the kitchen counter, on the floor, and in the medicine cabinet. I can't walk around without stepping on some. Suppose that I decide to give a candy bar to my friend, frequent Catallarchy commenter, and chocolate lover "Spoonie Luv". Is the value that he receives in the form of the candy bar the same as the value I lose in giving up the candy bar?

ticket.GIF• Suppose that instead of being a world-famous blogger, I am merely the owner of a movie theater. I need to hire a temp worker to do some filing in the office. Most temp workers doing the same type of work in the area are being paid $7 an hour over a forty hour workweek, or $280 a week. In an attempt to save money, I negotiate with the person I want to hire, and we come to a mutual agreeement that I will pay him $260 a week plus 10 free movie tickets a week, with the caveat that the free movie tickets can only be used during offpeak matinee hours. Is the value of the movie tickets that I give to him the same for me as it is for him?
-----------------------------------------------------
Edit September 28, 2003 - In these questions, I have inadvertantly asked to make an interpersonal utility comparison, while ironically arguing against being able to do such a thing in the comments. My intention was rather to show that the idea that some sort of Law of Conservation of Value exists is false.

Related posts:
Stock options compensation and opportunity cost
Employee compensation: cash vs. stock
Stock options compensation and opportunity cost II
Kling on stock-based compensation

Share this

In the instant of a

In the instant of a voluntary economic exchange, it may be inferred that each participant in the exchange subjectively values the good received more than the good given up. Except for this pair of reverse order comparisons, no other valid inference about any valuation at any time can be made.

In particular, subjective valuations for one individual are not subject to addition or subtraction. They can only be ranked on his single scale of subjective values at one moment in time.

Nothing at all can be said about the relationship of subjective valuations between separate individuals.

Exceptions?

1. Multiple owned samples of the same good subject to the Law of Diminishing Marginal Utility.

2. other?

Regards, Don

I understand the Austrian

I understand the Austrian attraction to subjective value, and it is an important point to make, but I still don't understand why Austrians refuse to add seperate individuals' values together. It may not always give us the perfect answer, but it is useful for any cost benefit analysis and is fairly reliable at that.

For example, we assume that a particular individual's willingess to pay is a good measure of the minimum subjective value that individual places on a particular good. So if I purchase a CD for $10, people who observe this purchase know that I value this CD at $10 or greater.

Thus, if we observe all of the individuals in a certain market purchasing multiple CDs for $10, we know that all of them get at least $10 of value, if not more. We might then try to estimate the demand curve and calculate the total social surplus above and beyond the given price of $10.

There can certainly be flaws in the estimation, but why would Austrians object to this practice in principle?

What you're asking is, "Are

What you're asking is, "Are there some goods that people value radically differently?" And yes, there are lots of examples of this. In fact, all transactions involve different valuations of the same good. Why else trade?

Micha, I don't think that is

Micha,

I don't think that is even a good approximation.

Money is an economic good, like all other economic goods. It is not an objective standard of measurement of value. $10 to you is very different from $10 to me is very different from $10 to Bill Gates is very different from $10 to a beggar.

Andrew, You're right, that's

Andrew,

You're right, that's exactly why we trade.

My question was not phrased very well, but essentially I was trying to show that in the context of options expensing:

1) There seems to be some kind of notion that if one party receives something of value, the giving party must lose an 'equal' amount of value. Losing a candy bar that I would probably eventually step on sooner or later would not bother me much; chocolate lover Spoonie Luv who can't get chocolates anywhere else would probably feast on it like a hungry wolf.

2) As Arnold Kling wrote on his blog in the link above, Warren Buffet says about stock compensation, "And if compensation is not an expense, what is it?" implying that compensation to one party *has* to be an expense to another party - a view that Arnold appears to endorse. Clearly, the expense to the theater owner of giving away tickets for seats that would likely go unused during offpeak matinee hours is essentially zero, yet is part of the compensation that the employee receives.

It may not be a "good"

It may not be a "good" approximation, but it is the best approximation we have. Our tools may not be perfect, but when making a decision, we must work with what we have. The alternative is to work with no valuation system at all.

And I agree that money is not an objective standard of measurement. Money is simply a convenient form of measurement because it is easier to count dollars and cents than it is to use "apples" as a form of measurement, convert everything in terms of apples, and then try to add up all of the apple values.

$10 to you is very different from $10 to me is very different from $10 to Bill Gates is very different from $10 to a beggar.

Sure, I agree with this, and economists, at least the one's that I speak to on a regular basis, frequently acknowledge our inability to make accurate interpersonal comparisons of value. But we can get a good estimate as to which one of various options will most increase the size of the pie, even if we can't determine the best way to divide the pie.

Clearly, the expense to the

Clearly, the expense to the theater owner of giving away tickets for seats that would likely go unused during offpeak matinee hours is essentially zero, yet is part of the compensation that the employee receives.

But there are only so many seats that the theater owner can give away. It is true that there is no variable cost associated with this particular action, because the theater owner is still operating between the threshold of two fixed costs (build one theater or two theaters.) The cost curve for this looks like stairs, because costs are constant until you get to a certain point, approach a large fixed cost, go back to being constant, and so on.

I'm not sure if the stock option case is entirely identical.

Regardless, each individual stock option granted is a lost opportunity, because although, theoretically, there is no limit to the number of possible shares, at a certain point, the current shareholders will put an end to any further dilution. In this sense, the potential number of shares to be granted is a limited, scarce resource and giving one to employees represents a lost opportunity to sell that same stock on the market.

Jonathan, "Suppose that I

Jonathan,

"Suppose that I decide to give a candy bar to my friend, frequent Catallarchy commenter, and chocolate lover "Spoonie Luv". Is the value that he receives in the form of the candy bar the same as the value I lose in giving up the candy bar?"

Well, in the end that is of course, totally subjective; additionally, Menger and two other braniacs independently figured out "diminishing marginal utility" which could be applied to the candy bar situation. You might, for whatever reason, value each one of those candy bars more than their weight in gold.

This can also be looked at through the supposed diamond/water paradox (use versus value). Ricardo and other's posited that diamonds were worth alot, but didn't have much use, water was just the opposite. However, he missed the larger picture in terms of scarcity and an individuals subjective preferences. You might, for whatever reason, value your candy bars more than the Hope Diamond.

Of course at that point, you'd be objectively crazy : )

"Most temp workers doing the same type of work in the area are being paid $7 an hour over a forty hour workweek, or $280 a week. In an attempt to save money, I negotiate with the person I want to hire, and we come to a mutual agreeement that I will pay him $260 a week plus 10 free movie tickets a week, with the caveat that the free movie tickets can only be used during offpeak matinee hours. Is the value of the movie tickets that I give to him the same for me as it is for him?"

Monetarily speaking, you both could possibly benefit. Now, to decide which one of you benefited "the most" is totally relative.

The reason why you could both benefit is that, first, you are saving $20 each week. The movies that your employee would attend would already be showing, so it's not costing you any additional money to satiate his taste. About the only thing I can think of that would cost you additional money as the employer is something like seat depreciation. If he decides to destroy the seats or cause unusual wear-or-tear, then you would have the short end of the stick (and if he repaid you, he would be losing cash too). Another way you could lose money is if he clogged a toilet which you had to hire a plumber to fix. Or if he burned out many lightbulbs by turning them off/on quite a bit.

In otherwords, for him to cost you any additional $$, you would need to assume that he doesn't do anything different than a "normal" and "civilized" patron would do (though, if employees get discount on anything, like corndogs, you could lose $20 eventually).

He on the other hand could be in a lose-win situation, because he might only be working at the theater as it is one of the few jobs that work around his school schedule -- he doesn't care about watching movies, nor would he watch them if he had the chance. So he'd be working the same amount of hours for less pay and monetarily gaining nothing. And I doubt he would agree to this deal - unless, you let him give/sell those 10 tickets, then he could be making money and we would be back to the scenario of "normal civilized" patrons.

Note: I have too much time on my hand.

Oh, I would like to point

Oh, I would like to point out I didn't explain "diminishing marginal utility."

Here's a quick example: you're hungry so you goto McDonald's and buy a burger. That fulfills X utils for you. Then you buy another, and that one fulfills Y utils for you (where Y is less than X). Then you get another one, and it fills you up Z utils (where Z is less than Y). This keeps going until you're no longer hungry and/or the additional burger just doesn't fulfill you anymore (whether it is some weird sexual fulfillment or simple biochemical energy).

http://www.wikipedia.org/wiki/Marginal_utility

With the chocolate bars, Jonathan could for some weird reason, value each bar as valuable as the next and therefore he doesn't want to give any of them up. Or he could care less as to what he does with a few here and a few there, so he throws them at hapless people below his apartment.

Of course, a question then could be asked, he values hitting Joe the Pedistrean more than his candy bars. How long until the marginal utility for that activity becomes zero?

Go Aggies!

Micha, "Regardless, each

Micha,

"Regardless, each individual stock option granted is a lost opportunity, because although, theoretically, there is no limit to the number of possible shares, at a certain point, the current shareholders will put an end to any further dilution. In this sense, the potential number of shares to be granted is a limited, scarce resource and giving one to employees represents a lost opportunity to sell that same stock on the market."

Economics, A Contemporary Introduction, 3rd Edition, 1994
by William A. McEachern, Professor of Economics, University of Connecticut, page 30,31 --

"Because of scarcity, whenever you make a choice, you must pass up other opportunities; you must incur an opportunity cost. The opportunity cost of the chosen item or the chosen activity is the benefit expected from the best alternative that is forgone."

"Opportunity cost is a subjective idea. Only the individual chooser can estimate the expected value of the best alternative. In fact, we seldom know the actual value of the forgone alternative, because by definition that opportunity lost is 'the road not taken' -- the alternative passed up in favor of the preferred option."

The question of whether stock or stock options are opportunity costs is apparently incredibly subtle, and depends on both characteristics of the grants and the precise definition of opportunity costs themselves.

Ideally, every management decision is based on maximizing shareholder value. At every instant in time, management has to make a decision as to whether either a stock grant or a sale of new shares is beneficial to shareholders. If both are beneficial, then the most beneficial is chosen first.

Both stock grants and stock sales are only beneficial in finite quantities. Stock sales only benefit the shareholders for as long the share sale proceeds can be re-invested by the company at a higher rate of return than the shareholders expect to receive from the company itself. Assuming a stock price valued on the NPV of the future cash flows to the shareholders, a stock sale that benefits the shareholders will increase the share price. Eventually, the next share to be sold will start to decrease the share price as the company can no longer find adequate investment rates of return for the received cash. Thus only a finite number of shares can be sold.

Stock grants also are limited in quantity as each successive grant uses up the amount of employee salary that can be beneficially substituted for with stock grants. The question of whether stock grants are beneficial is also determined by the rate of return that can be realized by the company for the salary cash freed up relative to the overall company rate of return expected by the shareholders.

Since both stock sales and stock grants are finite, they can be supplied by new shares even if the possible supply of new shares were not truly infinite. There just have to be enough.

Every possible beneficial share grant and every possible beneficial share sale is executed, alternating if necessary to execute the more beneficial first. Since an opportunity cost must involve a second best choice forgone, the fact that no second best beneficial choices can remain unexecuted means that no opportunity cost is possible.

The thing to remember is that just because the company can generate cash for a sale or a grant doesn't mean that it benefits shareholders to do so.

Regards, Don

Where's my KitKat? In the

Where's my KitKat?

In the movie ticket scenario, the worker may also benefit by having to pay less taxes. The theatre owner may lose more money if the worker gives/sells the tickets to moviegoers that would have originally paid for the tickets.

Well, not necessarily if the

Well, not necessarily if the tickets can only be used at off-peak times. I believe the Volokh conspiracy was discussing why movie theaters don't price discriminate a few weeks ago, but the movie owner in any case wouldn't really lose money- the ticket was "paid for" by the employee's labor (reversing the order of the equation, the employee bought $260 + 2 movie tickets in exchange for 40 hours of labor), meaning the employer explicitly shaved possible revenue off an off-peak showing to pay for needed labor, so he can't "lose money" twice on the deal.

I'm probably being naive,

I'm probably being naive, but I understand the whole issue with options thus:

Management is entering a contract to dilute current shareholder equity (whether by issuing new shares or taking shares from a VC as per Arnold Kling's scenario. Because the options may only be exercized at some point in the future, it is not always clear to current shareholders the extent to which their value is diluted and when this happens. Current shareholders want to make this more transparent.

Mark

The Austrian argument would

The Austrian argument would be that it is not even an approximation, and that it has zero validity.

That's an argument, but what decision criteria do they suggest should be used instead? Business managers clearly find econometric demand curve estimations useful when making business decisions about the nature of the market and consumer reactions to price changes. Do Austrians suggest that these managers are using invalid tools? If so, why do we see so many profit-maximizing businesses spending scarce resources on these market estimations if they have no validity? It seems to me that over time such wasteful methods would be weeded out.

The whole foundation of praxeology is in individual states of satisfactions that can't be measured.

But it is self-evident that we know the minimum value consumers place on certain goods at certain times. If we assume that consumer preferences do not fluctuate dramatically in very short time periods, and we are able to measure the minimum values of a large group of consumers at different price levels, why is it invalid to take this data and estimate a demand curve from it? True, this demand curve is far from perfect, and may change as time passes, but it seems like a useful tool for economic decision making.

Thus, the whole extension of Austrian theory is based only on relative *individual* states of satisfaction, and even then, they cannot be quantified and change moment to moment.

I don't see any reason to assume that consumer preferences fluctuate so dramatically over short periods of time as to be useless for decision making purposes.

You probably knew all that, and I suspect your question is more along the lines of "Why are Austrians so rigid in their methodology?"

I have no good answer other than they just are, although maybe I can try a roundabout way to answer. The problem with economics is that it is not a science that can be studied with controlled experiments. People make observations of the world and like to link events all the time. The people over at Atrios see poverty in the 3rd world and blame it on corporations. People at FreeRepublic see drug use rise in the 90s and blame it on Bill Clinton. This guy blames the boom and bust of the 90's on private industry while Austrians blame the Federal Reserve. Who's right and how do we know?

Well, I agree that economics and other social sciences in general (political science, sociology, etc.) do not have the "hard science" advantage of being able to conduct controlled experiments, but we are able to make observations under various conditions, build models and hypothesis based on these observations, and then measure the accuracy and validity of these models based on predictions about the future. Over time, and as we gather more evidence, we can build confidence that our models are correct or incorrect. Perhaps we can't have the same level of certainty as medical researchers, but such is life.

Mainstream economists agree on much, just as mainstream scientists agree on much, and there are minority opinions in both disciplines that reject the "accepted wisdom" of the majority. Neither Atrios nor the people on FreeRepublic are economists, and as for Quiggin, I think he has a point. It is clear to me that private investors made poor decisions about the nature of the Internet and the computer industry as a whole. The Federal Reserve may also have contributed to this failure. Both Quiggin and the Austrians may be partially right.

Many of the disagreements between left-liberal economists like Quiggin, Delong, and Krugman and right-liberal economists like Kling, Friedman (both father and son), etc. result from differences in subjective values regarding income and wealth distribution and use of natural resources.

So Austrians of the Misesian school study economics in a different way - by starting from primary axioms and building by deduction. They only believe what the praxeology tells them. If it can't be proven by praxeology, the don't believe it.

And does everyone agree with these axioms? I know that many philosophers reject the existence of synthetic a priori axioms. Do Austrians compare the results of praxeology with empirical evidence to see if their predictions accord with reality?

I'm already sympathetic to Austrian economics because I agree with many of their conclusions, but I have not yet studied in depth their methodology. It seems to me that even if I believe Austrian axioms to be correct, that won't get me very far in arguing with those who don't accept those axioms. In such situations, empirical evidence may be more useful.

It may not be a "good"

It may not be a "good" approximation, but it is the best approximation we have. Our tools may not be perfect, but when making a decision, we must work with what we have. The alternative is to work with no valuation system at all. [...]

Sure, I agree with this, and economists, at least the one's that I speak to on a regular basis, frequently acknowledge our inability to make accurate interpersonal comparisons of value. But we can get a good estimate as to which one of various options will most increase the size of the pie, even if we can't determine the best way to divide the pie.

The Austrian argument would be that it is not even an approximation, and that it has zero validity. The whole foundation of praxeology is in individual states of satisfactions that can't be measured. How satisfied am I at any given time? How satisfied are you? How satisfied is Kim Jong-Il? Thus, the whole extension of Austrian theory is based only on relative *individual* states of satisfaction, and even then, they cannot be quantified and change moment to moment. The only thing that can be concluded is that an action reveals preferences for the pursued state over the existing state. Value is thus imputed to means based on how they are used to pursue end states. In the aggregate, the values of multiple individuals result in the supply and demand curves of goods, which themselves are transient and change moment to moment. There is no underlying means, not even money, that can be used to measure the value to exchanged goods *across* individuals.

You probably knew all that, and I suspect your question is more along the lines of "Why are Austrians so rigid in their methodology?"

I have no good answer other than they just are, although maybe I can try a roundabout way to answer. The problem with economics is that it is not a science that can be studied with controlled experiments. People make observations of the world and like to link events all the time. The people over at Atrios see poverty in the 3rd world and blame it on corporations. People at FreeRepublic see drug use rise in the 90s and blame it on Bill Clinton. This guy blames the boom and bust of the 90's on private industry while Austrians blame the Federal Reserve. Who's right and how do we know?

My field of medicine has its very early roots in methods and remedies that has no basis in reality, yet people continued to perform bloodletting, exorcisms, homeopathy, etc into the 19th century. It was only the popularization of the scientific method that gave medicine a modern face. Take 50 sick people and perform bloodletting on them. Take 50 other sick people and do nothing. Is there a difference in the two groups? No!

Most likely, medicine prior to the 20th century was largely effective due to the placebo effect which still isn't explained well today. When I try to discern reality, when I try to see if a particular drug 'works', I ask myself, "Is there a double-blinded, randomized, placebo-controlled, large number study published about the effects of that drug? Have others confirmed those findings in other such studies on other populations?" If the answer is no, I am skeptical of the efficacy of that drug. It doesn't matter that the drug mechanisms have been studied in petri dishes, or that it has been studied in rats, or that the leading 'authorities' are espousing it. Until I see multiple, double-blinded, randomized, placebo-controlled, studies with statistically sound results, I remain skeptical. That is the only way to separate reality from gibberish.

However, economics cannot be subjected to controlled trials. I can't imagine how a double-blinded, randomized, controlled study in economics would be designed. The biggest limiting factor is the lack of a control - how do control for time? You can't.

So Austrians of the Misesian school study economics in a different way - by starting from primary axioms and building by deduction. They only believe what the praxeology tells them. If it can't be proven by praxeology, the don't believe it. Since there is no way to compare interpersonal utility based on praxeological deduction, Austrian don't do it. It is like my refusing to believe in the efficacy of a drug without scientific proof. Mises spends 200 pages in Human Action emphasizing the very rigorous nature of this method before even mentioning money or supply or demand. Any a priori methodology has to be rigorous, consistent, and principled to be valid. That is the Austrian way of separating reality from gibberish.

But there are only so many

But there are only so many seats that the theater owner can give away. It is true that there is no variable cost associated with this particular action, because the theater owner is still operating between the threshold of two fixed costs (build one theater or two theaters.) The cost curve for this looks like stairs, because costs are constant until you get to a certain point, approach a large fixed cost, go back to being constant, and so on.

Right, but at the time of the hiring of the temp worker, at offpeak matinee hours, many seats are simply not being used. It costs me zero to give 10 tickets as compensation, yet the worker receives compensation. I save cash by doing so, which can be 'reinvested' if I choose into my business. There is no reason this should be counted as an expense. If anything, my business benefited. You're right - in the limit, i.e., if I choose to do the same thing with many more workers, the finite number of seats will eventually cause problems, but that has to be evaluated on a case by case basis.

I'm not sure if the stock option case is entirely identical.

No, I don't think it is, and thus I originally stated that the examples may be analagous to a limited degree. My main point was that just because something is used as compensation, it does not necessarily result in an expense to the company.

Regardless, each individual stock option granted is a lost opportunity, because although, theoretically, there is no limit to the number of possible shares, at a certain point, the current shareholders will put an end to any further dilution. In this sense, the potential number of shares to be granted is a limited, scarce resource and giving one to employees represents a lost opportunity to sell that same stock on the market.

Either way, though, cash is raised. If I use stock as compensation, I save cash that would ordinarily have gone to compensation. If I sell stock, I raise cash. The shareholders lose due to share dilution. I'm not sure how this all plays out in opportunity cost, although the precise definition used matters a lot, and I am still developing my thoughts on the whole thing. However, I still cannot see how stock grants are an expense to the company. They dilute the shares and are an expense to the shareholders, but the company raises cash.

The reason why you could

The reason why you could both benefit is that, first, you are saving $20 each week. The movies that your employee would attend would already be showing, so it's not costing you any additional money to satiate his taste. About the only thing I can think of that would cost you additional money as the employer is something like seat depreciation. If he decides to destroy the seats or cause unusual wear-or-tear, then you would have the short end of the stick (and if he repaid you, he would be losing cash too). Another way you could lose money is if he clogged a toilet which you had to hire a plumber to fix. Or if he burned out many lightbulbs by turning them off/on quite a bit.

In otherwords, for him to cost you any additional $$, you would need to assume that he doesn't do anything different than a "normal" and "civilized" patron would do (though, if employees get discount on anything, like corndogs, you could lose $20 eventually).

He on the other hand could be in a lose-win situation, because he might only be working at the theater as it is one of the few jobs that work around his school schedule -- he doesn't care about watching movies, nor would he watch them if he had the chance. So he'd be working the same amount of hours for less pay and monetarily gaining nothing. And I doubt he would agree to this deal - unless, you let him give/sell those 10 tickets, then he could be making money and we would be back to the scenario of "normal civilized" patrons.

Yes, this is exactly what I was trying to say. Using different types of compensation can result in a benefit to both the company and the employee. Compensation doesn't necessarily imply expense, and non-cash compensation can actually raise cash for the business.