This post is the second in a series of articles that explains the new economic theory I proposed in Five Short Blasts. If you haven’t read Part 1 yet, just go to “The Theory Explained” category of this web site. This series of articles will be archived there in reverse chronological order. Just scroll down to find the beginning of the series.
* * * * *
In Part 1, we boiled down GDP, economists’ favorite measurement of economic growth, and found that every bit of the growth in per capita chained GDP is accounted for by gains in productivity. Only if those gains are accompanied by real increases in income (adjusted for inflation), as politicians, business leaders and economists say they are, will our standard of living rise. But I ended Part 1 by claiming that this is not true. There is no correlation between productivity improvement and wages. It’s something else that drives increases in income. In this article, I’ll explain.
The U.S. Department of Labor’s Bureau of Labor Statistics in 1987 began publishing compensation data for hundreds of categories of labor involved in a wide range of industrial activities like manufacturing, wholesaling and retailing, among others. They report on compensation, productivity, hours worked, and so on. I plotted the data for compensation vs. productivity as shown in Figure 1-5.
I was quite surprised at what I found. There was no correlation whatsoever. None. The chart looked like a shotgun pattern. Increases in productivity were no more likely to produce a gain in income than a decrease, and vice versa. In fact, the biggest gain in productivity since 1987, in NAICS code 3341, the manufacturing of computer equipment, a productivity gain of 3,800%, actually yielded a 24% decline in compensation.
It makes sense when you think about it. If productivity is improved and a product can be made with less labor, what is the employer’s motivation to pay his workers more? His need for labor just went down. He just tossed some workers back into the labor pool, increasing the competition for jobs. He can now get away with paying them less, not more.
Does that mean that productivity improvement is a bad thing? Not at all. It’s human nature to find ways to accomplish work more efficiently. It’s productivity improvement that keeps companies competitive with each other, and it keeps our nation competitive with others in global trade. It allows for an improved standard of living by freeing up labor to work on unmet needs and challenges. But, clearly, it’s not what drives incomes higher. There must be something else.
But as I compiled the data for Figure 1-5, I noticed what seemed to be a better correlation between compensation and another variable – total hours worked. So I decided to try a plot of compensation vs. percent change in total hours worked. The result is Figure 1-6.
Again I was surprised, not so much that there was a correlation but at how strong that correlation was. It’s rare to see such a powerful relationship. But when you think about it, it makes sense. The percent change in total hours worked is simply a measure of the demand for labor. If it’s rising, then employers must compete for those workers and offer them higher wages to attract and keep them. If the demand for labor is falling, then there will be an overabundance of those workers and employers can pay them less. It’s the old law of “supply and demand” at work, the most basic precept of economics.
Speaking of “supply and demand,” we can’t forget about the “supply” side of that equation. I need to modify my above conclusion. If the demand for labor is growing faster than the supply, only then will an increase in compensation be the result. If the demand for labor grows, but not as fast as the supply, then we will actually see compensation slowly decline.
Now the question becomes, “what is it that drives the demand for labor?” It’s people who consume workers’ production. If they want more of that particular product, then the demand for labor to produce that product will go up. If we double the number of people who use that product, then the demand for labor to produce that product will also double.
Or will it? We have to remember that workers and consumers all come from the same pool of people. If we double the population, then not only have we doubled the number of consumers but we’ve also doubled the number of laborers too, more or less. What happens if the ratio doesn’t stay the same? What if consumption, for some reason, doesn’t keep pace with the growth in the number of workers? Then the demand for labor needed to produce that particular product will decline. If there isn’t a corresponding increase in the demand for workers to make some other product, then the overall demand for labor across the total spectrum of products will decline, which will begin to drive down wages for everyone. That would be a very bad thing with lots of bad consequences – rising unemployment and poverty, and all of the other side effects that accompany them, things like increased social spending by government (funded by higher taxes), a rising crime rate, and so on.
The only meaningful way to gauge how we’re doing in the “supply vs. demand” equation for labor is to look at everything in per capita terms; that is, we need to compare how the average person’s consumption of products stacks up against his average contribution to the labor force. If that ratio begins to decline, we’ve got a problem. If, on average, people begin to consume less while their per capita output (another way of saying “productivity”) rises, then we have an economically unsustainable situation.
Let’s consider an example. Suppose we have a population of 1,000 people and, on average, each person consumes one widget per year. Five hundred of these people are workers, while the other five hundred stay at home for various reasons. The productivity of these workers is such that each worker is able to produce two widgets per year. Supply and demand are in perfect balance. 1,000 widgets produced. 1,000 widgets consumed.
Now, suppose that we add another 1,000 people to the population, including the addition of 500 more workers but, for whatever reason, these added people are only able to consume one half of a widget per person on average. What will happen? Well, total widget demand will rise to 1,500 widgets. If you’re the CEO of The Widget Company, you’re very pleased with this huge increase in sales. But what if you’re one of the workers? The potential output of the workers has risen to 2,000 widgets, while the demand has risen to only 1,500. What will happen is that 25% of these workers will be out of a job. Again, the CEO of The Widget Company likes this situation. The oversupply of labor means he can pay less and make more profit on each widget he sells.
In this example, the per capita consumption of widgets fell from 1.0 to 0.75, while the per capita output (or productivity) held steady at 1.0 (2.0 per worker, which is 1.0 “per capita”). Per capita consumption declined while productivity remained unchanged. The result was rising unemployment.
This is why it’s important to track everything in per capita terms, both per capita consumption and the per capita demand for labor. It’s this balance that sustains our standard of living. If it gets out of balance either way, it can raise or reduce our standard of living. Now, throughout nearly all of human history, if productivity improvement freed up workers in one field, there was another product need that was going unmet due to a lack of workers. When cave men pooled their resources and became more efficient at hunting and gathering, someone was freed up to begin inventing the wheel. When they became more efficient at making wheels, then someone was freed up to begin making copper tools, and so on.
But what would happen if that reached a limit? What if per capita consumption couldn’t keep pace with the rise in productivity? Worse still, what if per capita consumption actually began to decline for some reason? We already know what would happen – rising unemployment and poverty, and everything that goes along with it. I think we can all agree that, from an economic perspective, anything that would drive down overall per capita consumption would be a bad thing, something to be avoided, something to be corrected if it took hold.
There is something that is doing exactly that, driving down per capita consumption, slowly but surely and dramatically, here in the United States and across the globe. In Part 3 we’ll find out what that is.