A commenter on my gestalt production post had a lengthy comment lost by the web form, and hasn’t had time to rewrite it. He did leave a short comment mentioning the supply and demand theory of pay. Jacob, I’m going to try to describe the theory I think you were referring to and then write a response. If I get it wrong please correct me. This grew too long to be a comment so I’m making it its own post.How are wages determined? According to traditional economic theory they are determined by the marginal revenue of labor. Essentially, if adding another worker will allow the company to make $50,000/year additional revenue then the company will make more money if they can hire that worker for any amount up to $50,000/year. This is the marginal revenue of that worker. In this context, marginal means the change in output due to a small change in input.
According to the law of diminishing returns, each additional worker will produce less marginal revenue than the one before. As a company hires more people, the marginal revenue of labor goes down. This creates a demand curve for labor. If a company currently has few workers it is willing to pay more because the marginal revenue of the next worker is high. Likewise, if a company has many workers it won’t be willing to pay as much because the marginal revenue of the next worker is lower.
The supply of labor is governed not only by the number of workers, but also by their relative preference for money vs. leisure. If jobs are only paying $6/hr maybe I’d prefer to just work part time, live in a hovel, and enjoy more leisure time. If jobs are paying $100/hr, the extra compensation will motivate me to work more hours. Basically, to get me to work an hour, the company has to pay me more than how much I value an hour of leisure. How much I value money vs. leisure also has diminishing returns so to get my first dollar to buy food I’m willing to work a lot, and to get my last hour of leisure a company would have to pay me a lot.
This creates the traditional supply/demand relationship where increasing wages create more labor supply and less labor demand. Where they meet is the equilibrium point. If a worker has scarce skills, or skills that can create more marginal revenue per hour, such as a CEO, then that affects the supply and demand curves for that type of worker and the equilibrium wage might be higher. It might be a lot higher, but there’s nothing wrong with that. Wages are always set by the impartial forces of supply and demand, and markets always operate at equilibrium so prices are always right.
Jacob, let me know if I got that wrong, or if there’s anything you’d like to add.
The above was my understanding of the traditional theory. Below is my response.
I’m not really going to address the issue of whether the supply and demand theory of pay is an accurate description of how wages work in our real economy. I think it has some problems like the difficulty of a firm actually being able to predict the marginal revenue from an employee, especially a unique high-profile employee like the CEO, whether workers are really as free to choose between work and leisure as the theory assumes, and frictional effects like CEOs being on each other’s boards of directors and tacitly agreeing to give each other higher wages.
But I’m not going to focus on that here. For now, let’s just assume that wages really are equal to a worker’s marginal revenue. My main point in the gestalt production post was that marginal revenue is not equal to how much each worker produced.
One idea that libertarians hold to be a self-evident truth is that every individual has an inalienable right to own the economic product of his or her individual efforts. A further idea that libertarians believe is that when you work for wages, whatever you are paid must automatically be equal to the economic product of your efforts. Finally, according to the supply and demand theory of pay, this is equal to the marginal revenue of your labor. What my gestalt production post was primarily about is that these three claims taken together produce a contradiction.
In general, there is no reason why summing the marginal revenue of each worker should add up to the total revenue of the firm. In fact, if you believe the supply and demand theory of pay I can prove that they must be unequal. The marginal revenue that determines the pay for all workers (or all workers of one type) is equal to the revenue change of adding or firing a single worker, the so-called “last worker”. But due to diminishing returns, the last worker must be the least productive. So the actual productivity of all workers of that type other than the last one is greater than their marginal revenue. Marginal revenue of the last worker times the number of workers must be less than summing the actual revenue over all workers.
So the firm will be left with a residue of funds, the difference between marginal revenue times number of workers vs. actual revenue. Capitalist theory takes a left turn at this point, and assumes that this must be the economic productivity of capital goods, the machines that the workers use. But why should we assume that these two quantities are equal? It sure is a convenient idea, but why should it be true? Wouldn’t it be fair to judge capital by the same standards as labor? Lets take each machine out of the process, one at a time, see how much less we would produce without that machine, and that’s the marginal revenue of that machine. The economic productivity of each machine, and how much profit the owner of that machine deserves, should be the marginal revenue of that machine, just like labor. Once again, there’s no reason why the sum of marginal revenue over all workers plus all machines should equal total revenue.
So what do I think I’ve proven? Libertarians believe the following two statements:
#1: In a group effort, the total economic product can be divided up into separate parcels where each parcel contains what was produced by the individual effort of one of the individuals in the group, and what each individual deserves to be paid is the value of their individual effort.
#2: In a group effort, what each individual deserves to be paid is the marginal revenue loss of removing that individual form the group.
What I’ve proven is that the two quantities, “the value of what was produced by an individual’s effort”, and “the marginal revenue loss of removing that individual from the group”, are in general not equivalent.Â Â If you take all of the parcels of individual production and add them all up they had better add up to the total production of the group.Â But if you add up the marginal revenue of all individuals, it won’t necessarily add up to the total revenue for the group.Â So the two quantities can’t be the same.
Therefore, these two statements are incompatible because they say these two unequal quantities are both equal to the quantity “what each individual deserves”. Capitalist theory has tricked us into thinking that they are the same so that the philosophy of “each individual deserves what they individually produce” can be used to justify the policy, “each individual receives the marginal revenue loss to the firm of firing them”
I haven’t proven that either of these ideas is wrong. I’ve just proven that they can’t both be right. Now I’ll shift gears a little and talk about these two ideas separately.
First let me say that these two statements are two value judgments. They both define what an individual in a group effort deserves. They are both definitions, not the consequence of a line of reasoning. These are not the only possible definitions of what an individual deserves. Fundamentally, when we agree with one of these it’s because we believe it is fair. Have a look at my posts on values and federalism to see how I believe we should handle these kinds of value questions. Basically, I think we should live and let live, and use federalism to have many different kinds of societies so people can choose to live in a society that matches their values. Here I’ll be telling you my values, but I don’t expect everyone to agree with me.
For #1, I think that in a group effort the quantity “what was produced by the individual effort of one of the individuals in the group” is undefined. Splitting the product up in this way is like dividing by zero. It can’t be done. There is no sensible definition of this quantity that guarantees that the sum of all the individual products is equal to the actual total product of the group. If someone could come up with a definition I would like to hear it.
I actually think the idea “each individual deserves what they individually produce” is morally acceptable. It’s not my most preferred definition, but I wouldn’t find fault in anyone who believed this. The problem is that it’s totally unusable in practice since almost everything is produced by group effort. Even if you go off into the woods by yourself there are still questions of equal opportunity. How much of your productive ability came from the fact that you grew up in a first world nation and got a good education as opposed to growing up in a third world nation having to struggle to avoid starvation? Wasn’t your production affected by the people who helped create the environment in which you find yourself? Is there really anything at all that you do where you can say it was totally unaffected by anyone else?
For #2, as far as I can tell, the only moral argument for this was the idea that marginal revenue is equal to what you produce, which has now been debunked. It’s a convenient and easy-to-implement system because you just have to allow anything-goes negotiation, but despotism is convenient and easy-to-implement too. That doesn’t make it fair. There is another argument for the idea that whatever you can negotiate is what you deserve, which is basically that no one owes anything to anyone. Life is the war of all against all so anything goes is fair. I don’t agree with that argument, and I wouldn’t respect someone who did.
What do I think is the definition of who deserves what? First of all, I don’t believe that everyone always deserves the same. There’s a liberal line of thought that no one can deserve more that someone else so any time one person has more than others it must be unfair. I totally disagree with that. If you work harder, you deserve more. What people deserve should be based on what they do and the choices they make.
I don’t have a hard and fast definition. I haven’t figured it out yet. But I do think that our common sense can serve as a “sanity test” to detect situations that are obviously unfair and where the system is obviously broken. For example, what should the ratio be between the salary of the best CEO in the world and an average worker? CEOs do work longer hours. Maybe the CEO works 120 hours per week (that’s 18 hours a day, 6 days a week plus 12 hours on Sunday) while the worker works 40. The CEO should earn 3x for that. And then there’s the amount of effort spent while working. There’s a difference between busting your butt to do the best you can and just punching your timecard. I’ll give another 3x for that. Then there’s the unpleasantness of the job. Sewer workers should be paid more than candy taste-testers. It would be hard to argue that flying around in corporate jets and staying in executive suites is unpleasant, but I suppose the job is stressful, and you have to be away from your family a lot on travel so 2x for that. Then there’s how skilled a person is at their work. This is an area open to some debate. If a person was born with talent do they deserve to be paid more when they didn’t do anything to “earn” their talent? And how much of being skilled is due to things under a person’s control as opposed to inborn talent? I won’t answer this question. I’ll just give the CEO 2x for being highly skilled instead of just an average manager. Then there’s the value of the work itself. I will step back and talk about that a little.
Different arrangements of labor in our workforce can produce different amounts of economic value. A factory making delicious cookies certainly produces more value than if those same people spent the same time and effort making mud pies. And even within an existing organization, changing from a bad manager to a good manager can increase the total output by more than changing from a bad delivery boy to a good delivery boy. I don’t think the manager can take credit for all that difference by himself as if the people he is managing had nothing to do with it. But I can’t deny that there are some positions that have more leverage on the total economic output of a group effort, and it’s more important to have the best people in those positions than in other positions.
Now I think a lot of the “value to society” of your job is outside your control. For example, I’m a computer programmer. There are many highly paid jobs programming computers nowadays because computers help produce a lot of economic value. I happen to have a talent for doing this so I get paid more because I work in this field. But I could easily have been born fifty years ago before these jobs existed. I could have wound up as a high school math teacher earning half or a third (inflation adjusted) of what I’m earning now. There’s no difference in my skills or how much effort I put into pursuing my career. Frankly, I don’t feel like I did anything to deserve this extra money. I was just lucky enough to be born at a time when the career that interests me and that I’m good at happens to have higher than average economic value.
But, there’s another argument that people might choose a job that produces more economic value over a job they would rather do that is less valuable. Maybe our CEO is a very talented manager, but loves music. He is sacrificing himself by working in a job that can produce more economic value to society rather than trying to be a concert bassoonist, which he would really prefer, but would result in less economic value being created. He should be compensated for his sacrifice.
And then there’s the argument that we need to pay more as a motivational incentive to get the best people in these positions separate from any argument that they deserve more. I don’t really know what the answer is, but I’ll give my CEO 5x for his job being a more valuable job.
So if you multiply these out you get: 3x3x2x2x5 = 180. I would have no problem if the best CEO in the world earned 180 times the median wage. The median wage in the US right now is around $45,000/yr. That would give the CEO $8.1M/yr. Not bad, probably higher than most liberals would feel is right, but much lower than a lot of current executives. And the analysis I did was for the best CEO in the world, not an average CEO. Maybe an average CEO works 80 hour weeks (2x), works 2x as hard as an average worker, still gets 2x for the stressful job, gets 1.5x for having above average management skills (average for a CEO), and still gets the 5x for the value of the job. That’s 2x2x2x1.5×5 = 60 or $2.7M. That’s what I think would be fair.
More important than the number is the idea that we need to look to external standards of fairness for why should someone deserve more than someone else. We can’t expect our economic system to tell us who deserves what. We need to decide on what factors make one person deserve more than another based on our values, and then make sure our economic system produces wages in line with that definition of fairness. Our current corporate capitalist system does not do a fair job for super-high earners.