The law of the decreasing marginal utility of income describes a universal fact of human experience, namely, that we tend to value the first dollars we earn more than the later ones. This is just another way of saying that we tend to purchase necessities before luxuries. Food and shelter are more important than fashionable clothes or toys for the children. In general, each dollar we earn is worth less to us, in subjective terms, than the dollar before. It is a special case of the general law of diminishing returns that is observed in connection with wide variety of physical and psychological phenomena.
The second law of economics Jevons appealed to was the law of the increasing marginal disutility of labor. In a way it is just the opposite of the first, a kind of mirror image describing the tendency we all feel for the first hours on the job to be easier than the later ones. As a general rule, the longer and harder we work, the greater our pain and sense of fatigue, a tendency that increases exponentially with time on the job and the intensity of our effort.
Why is this so? Well, it clearly has to do with our physiological nature, which makes fatigue an accumulative phenomenon and a progressively accumulative one at that. This is a phenomenon everyone is familiar with from first-hand experience.
Now it was Jevons's particular stroke of genius to plot these two empirical laws – one a law of subjective psychology, the other a law of physiology -- on a single graph measuring utility against time. In this way he was able to compare the marginal utility of wages with the marginal disutility of labor for each successive hour on the job, as we see in Figure 4.
Jevons saw that there was a point, marked on the graph at point t, where the utility of wages just balances the disutility of labor. He reasoned that a worker would voluntarily continue laboring at a given pace only to that point, beyond which the marginal disutility of his labor would exceed the marginal utility of his income (which is also the point of maximum total satisfaction). Beyond point t he would begin to slack off in order to ease his pain and bring it into line with the decreasing marginal utility of the extra money he was earning.
Moreover, since a worker’s sense of fatigue is an exponentially increasing phenomenon, he would tend to slack off more and more, eventually coming to a complete halt. His employer might manage to delay or otherwise mitigate this natural tendency to slack off, but only by threatening his employee with dire consequences, such as docking his pay for the whole day or firing him from his job.
This is the situation that has always prevailed in the West ever since the introduction of the factory system at the beginning of the Industrial Revolution.
But let us now apply these analytical tools developed by Jevons to the situation at hand. Before doing so, however, I want to note three tacit assumptions that were made by Jevons at the time he formulated his theory of labor: First, he assumed that the length of the working day was fixed by custom. Second, he assumed that the pace of work was likewise fixed by custom. And third, he assumed that each worker would receive a fixed hourly wage which his employer had agreed to pay him, and which would not vary according to circumstances. It is hardly surprising that Jevons made these assumptions; all three of them, after all, were firmly established by the customs of his age, as in fact they continue to be established by custom in most industries to this day.
Yet when we relax these three assumptions, a wholly different picture begins to emerge: a picture whose remarkable implications are not at all obvious, to judge by the way we do things today.
In the first place, if we cut the length of the working day in half -- which is essentially what we are proposing – then we shall, in effect, have moved our future part-time workers well to the left of point t in the diagram, where the marginal utility of their incomes greatly exceeds the marginal disutility of their labor. All else being equal, this means that their total job satisfaction will be less than before. They will not be happy with so few hours.
But, of course, all else will not be equal. For one thing, the pace of work on the factory floor will no longer be fixed by custom. Rather, it will be determined by the workers themselves, at whatever intensity they desire. Furthermore, they will no longer be remunerated at a fixed hourly rate. Instead each worker’s pay will be based on his fraction of whatever he and his fellows are able to produce collectively in a given period of time: labor’s total share being calculated by a formula which they shall have negotiated with their employer ahead of time, which is to say, which is to say, on the basis of a formula which shall have been established through a process of collective bargaining.
What this means is that our part-time workers will no longer be stranded to the left of point t. Instead they will now have both an incentive and the opportunity to work much harder and more efficiently than they have in the past, which will move them to the right on the diagram, bringing them closer that natural point of equilibrium where their total satisfaction is at its theoretical maximum.
Thus we arrive at the same conclusion by means of this Jevons diagram as we reached through simple intuition. Factory workers can rationally be expected to work faster and more efficiently if their hours are reduced and their pay is tied to their output – assuming, of course, that this increase in their productivity is shared equitably between labor and management. By equitably I mean that the percentage increase in real hourly wages should be the same as the percentage increase in the rate of return on investment.
Employers might also take a moment to ponder the issue of worker morale. One of the ironies of the contemporary workplace is that employees often seem least happy in precisely those industries that have the best pay and benefits. It is not in sweat shops but in these seeming bastions of industrial privilege that witness the unmistakable signs of worker alienation: high rates of absenteeism, tardiness, goofing off on the job, even occasional acts of vandalism and theft. Why should this be?
As should be clear by now, the most parsimonious explanation is that people employed in high-wage industries are more likely to be working beyond the point of their natural equilibrium. Or in the language of Jevons, they are more likely find themselves to the right of point t where the marginal disutility of their labor exceeds the marginal utility of their incomes. Of course many of these workers may not “know’ that this is the cause of their discontent, but intuitively they feel it, and they respond accordingly: by slacking their pace, goofing off, and not always showing up to work on time. The luxuries that they are able to purchase with their paychecks – bass boats, off-road vehicles, and the rest – do not compensate for those extra hours on the job, even though they may not care to admit it. The symptoms are unmistakable.
Contrast this to what you are likely to see in the case our proposed part-time workforce. Needing every dollar they earn to support their new leisured lifestyle, the marginal utility of their incomes will remain very high. This isn’t to say that these workers won't be able to afford any luxuries. But it does mean that the luxuries they do purchase will tend to be more modest. Indeed, in a sense, their one big luxury will be the new lifestyle itself -- which, being an all-or-nothing proposition, will be something like a luxury and a necessity at one and the same time. After all, how does one put a price on one's whole way of life? It stands to reason, therefore, that the members of this new workforce, unless prevented from doing so by real sickness or injury, are going show up for work without fail, ready and eager to give it their all.
Let us now get down to brass tacks. Just how much harder and more efficiently can we expect these new part-time workers to work? How big of an increase in output-per-man-hour are they likely to show?
This is, of course, the sixty-four dollar question. It also happens to be an empirical question, which no amount of econometric calculation can estimate, not even approximately. To make matters worse, there does not appear to be a great deal of empirical evidence out there that bears on the issue. But there is some evidence, so let me summarize the little I have been able to find in the past forty years, including the results of my own little experiment as a small-time employer of a part-time workforce.
To date I have managed to identify three major employers in the United States who have hit upon the idea of combining a shorter workweek with an incentive-based wage system. The first and in many ways the most impressive is The Lincoln Electric Company, an old and highly-successful manufacturer of welding equipment that is based in Cleveland, Ohio. It has long maintained a generous profit-sharing arrangement in conjunction with a piece-rate system, which it combines with a flexible workweek of around 35 hours and a policy of not laying off workers during a downturn.*
The second company is United Parcel Service, the well-known package delivery company. UPS has long been famous for its brown trucks, high wages, and the hustle of its drivers. In the past in some of its regional sorting centers in the Mid West – centers which just happened to be located in rural areas-- employed teams of locals to sort packages during the night. There was an agreement that everyone would get paid the same amount money to sort the night’s packages no matter how quickly or slowly they accomplished the task. With this understanding, it typically took the team in the center three or four hours to do the job, which meant that the individual team members were getting an unusually high wage when measured on an hourly basis. This is something I learned about through personal conversations with company officials a number of years ago. I do not know if they have continued to maintain this arrangement.
The third company I have been able to identify is SteelCase Inc., the Michigan manufacturer of office furnishings. Like Lincoln Electric, SteelCase uses a combination of profit sharing and work sharing, and has had similar results.
Without going into the details let me simply state that all three of these companies has managed to wring a 30-to-40 percent increase in worker productivity. That is to say, by combining shorter workweeks with group-incentive wage systems, they have managed to consistently maintain levels of productivity, wages, and return on investment that are 30-to-40 percent higher than those of their competitors. What’s more, as anyone who is familiar with these companies knows, all three of them dominate the markets they are in.
Perhaps not uncoincidentally, this is broadly consistent with my own experience as a private employer. For a quarter of a century my wife and I operated a small landscaping company out of our house. When we first started out in business, my wife Pat was already a trained horticulturist as well as a talented garden designer. I, on the other hand, was a journeyman carpenter, having entered that profession after graduating from college, my intention being to find out whether it would indeed be feasible for ordinary people to build their own houses. (I determined that it would be feasible with a certain amount of professional assistance.) In any event I was already used to physical labor. Thus our plan was that Pat would bring in the jobs, while I and a third partner in the business, a Greek stand-up comedian named Craig Devoulu,s would be responsible for getting the plants installed. To help us we hired a common laborer named Joel Arwood, a man so honest and true he could have been Jesus come back to earth for all I could tell.
Thanks to the talents of my wife, our little company prospered from the start. But after a couple of years our third partner bowed out (it wasn’t easy being in partnership with a man and his wife) at which point I hired two more common laborers. It was at this point that I took opportunity to test out my little pet theory of labor, which I had already been mentally toying with for a number of years.
My wife was naturally dubious about the whole idea. But I pointed out that we were already working part-time (thanks to inclement weather, and the seasonal nature of our industry) and that we could begin by paying our workers the same share of the proceeds they were already getting, on average, with their current hourly wages. Thus I was able to assure her that if our employees didn't work harder or more efficiently than before they wouldn't make a penny more, and nobody would be the worse for it. So she reluctantly consented to give my little experiment a try.
Several days later I sat down with my crew, job sheets in hand, and went over the last dozen installations we had done together. On the front of each sheet was a carbon copy of the contract that had been signed by the customer, listing all the plants, soil, stone, mulches, and other materials that were scheduled to be installed, and the total price which the customer had agreed to pay. On the back of the same sheet was a list of the wholesale prices we had paid for all these materials, with the supporting receipts, and a record of each worker's hours on that job and his individual wage rate, my own included.
With this information in hand, it was a simple matter to calculate the net proceeds of each job and labor’s collective share of those proceeds, on average, over the past several months. The net proceeds -- or the ‘share of the net’ as I called it -- was defined as the difference between the contract price of the job and the total cost of the materials used.
So I asked my crew what they thought of the idea that, on each future job, I simply pay them the same percentage share of the “net” as they had been receiving, on average, over the past dozen jobs? That way if they completed a job more quickly they would be able to earn the same amount of money in a shorter period of time. Seeing the logic of what I was trying to say, and of course the possibility of upping their paychecks at the end of each pay period, they agreed to give it a try.
What happened next? Well, what happened next is that starting with the very next day I witnessed a 40% increase in worker productivity: an increase that remained more-or-less constant through the rest of that season and over the course of the next twenty-odd years.
The key to my success, I think, was my decision to sit down with my workers at the end of each job, with the customer's check in one hand and a fist full of receipts in the other. We would then proceed to fill out the back of the job sheet that went with each contract. There were no secrets here and everybody knew it. They got 29% of the difference between the price of the job and the cost of materials. Furthermore, if I made a mistake in arithmetic as I occasionally did, nobody was shy about pointing it out. As long as my workers knew they could trust me, that I wasn't trying to put one over on them, they were fully on board with this new arrangement. And why shouldn't they be? They were now taking home on average forty percent more than they had before, which was forty percent more than a lot of their friends were making, who worked for my competitors.
I also discovered that there were some fringe benefits to this new arrangement which I had not anticipated. For one thing, it didn't seem to matter anymore whether I was on the job or not. It was no longer necessary for me to keep an eye on my employees to make sure that they were working and not resting on their shovels, which was something I had always hated doing. And when I eventually hired an additional employee (our little company was growing) and he started goofing off the first time I went away from the job, his co-workers quickly put him in his place. Everybody was working for each other now, as they well understood, and they made sure the newcomer understood it as well.
Something else I noticed was that the waste of materials dropped off significantly. There were fewer damaged root-balls and broken limbs than in the past, fewer poorly planted shrubs and trees that would need to be planted all over again. Plus I discovered that my gardeners (they were no longer common laborers) were now paying close attention to the finer points of arranging the plants on the ground, in accordance with the designs of my wife. In short I was now seeing less of everything which could possibly reduce the size of their paychecks at the end of the week. I still had to monitor the quality of their work – to make sure they did not skimp on materials -- but my wife and I were both gratified to find that we had been relieved of many of the burdens of managing a crew.
We were also pleased to see a 40% increase in our own earnings when calculated on an hourly basis. My wife would occasionally question the logic of paying our workers so much more per hour than our competitors did, especially when they were installing a particularly “fat” job with unusually high profit margins. I would tell her not to worry. The more money they were making the more money we made. W e both should be thankful for the fact that our little company no longer suffered many of the headaches endemic to our trade. In particular we no longer had problems of chronic absenteeism and high rates of employee turnover, nor were we in the habit of losing our most highly-trained plantsmen to our competitors. These were all things that had caused not a few of our competitors to quit the business altogether over the years.
To sum up, based upon my own personal experience as well as the experience of UPS, SteelCase, and Lincoln Electric, I predict that manufacturers who invest in our plan will see increases in worker productivity on the order of 30-to-40 percent, depending upon the particular circumstances of the manufacturing process involved.
Of course such productivity increases are subject to proviso which I have already identified, namely, that they be shared equitably between labor and capital according to a mutually agreed upon formula that seems fair to both sides. As for the best way to arrive at such a formula, let me just say that this is a difficult issue which must be approached very carefully.
Some situations, for example, are relatively straightforward. Adjusting the speed of an assembly line presents no particular problem because any given percentage increase can finance the same percentage increase in wages and in rate of return on investment. But in other cases this will not be so. What happens when we encounter a situation (in a carpet mill for example) in which just a few employees operate a large number of very expensive machines whose speeds are not adjustable without a radical redesign? Even if by working more diligently half as many workers are able to operate the same number of machines, they should not expect to double their wages; doing so would leave nothing for management. Thus not every production process lends itself equally to incentive-based work sprints, a point that needs to be made. We are going to have to deal with cases one at a time.
But in every case, let me re-emphasize, some form of collective bargaining is going to be required. There should be no misunderstanding on this point. For that reason I am attaching as a technical appendix a note on wages and prices that I wrote several decades ago. Whether this note will be of any practical use, or even makes sense, I can no longer tell. I will leave it to others more qualified than I to decide.*
Let me conclude this discussion by returning again to the question of the expected new standard of living. At the end of the last chapter I listed five different ways families would be able to economize in their new family budget. Nevertheless I questioned whether all of these savings together could possibly compensate for such a large reduction in hours on the job? I admitted that skepticism would be in order.
We are in a much better position now to address such skepticism. I have just shown why the actual loss in wages might not be nearly as large as one might have initially supposed. Certainly the loss will be less, in percentage terms, than the actual reduction in the total hours of employment. In fact for a certain class of families -- those with less than two full-time wage-earners – there may be no loss of income at all. Families that still get by on the earnings of a single full-time worker might actually see their incomes go up.
But let me stick to the case of a young married couple who are the parents of two pre-school age children. As is often the case nowadays, I will assume that both parents are forced to work full-time outside the home in order to support their family. How might such a couple fare in terms of our plan? Given that they are currently working for wages 80 hours a week, if they switch to two 18-hour-a-week jobs, assuming they sustain a 40% increase wages, their after-tax income might go down by less than one-third, notwithstanding that their total hours of employment had fallen by more than half. More realistically, let us assume that they switch to two 24-hour-a-week jobs outside the home and are able to sustain only a 20% increase in productivity. In that case their after-tax income would shrink by one-forth.
Clearly, these calculations lead to a more optimistic assessment of what the family’s future standard of living might be: -- once it no longer has to purchase and operate two full-sized automobiles, can stop paying for daycare, starts cooking at home more often and eating out less, and no longer has to set aside large sums to pay the future costs of retirement and nursing-home care.
At the end of the next chapter I shall conclude my ongoing analysis of home economics by estimating the amount of money a couple might save by building its own house in the countryside. Depending upon the choices they make, it could be a lot more than you imagine.