That’s one issue. Two others are “stretch” and cyclicality. Together they explain why manufacturing wages haven’t caught up with the demand for labor. What happens when they do?

In North Dakota, the oil boom has driven labor rates so high that a Wal-Mart offered starting wages of $17 an hour. Would an established manufacturer in this state—say, one making office furniture or aftermarket auto parts—be able to raise its employees’ wages to the same extent?

Probably not. The store serves local customers in the local economy. By contrast, the manufacturer almost certainly ships to (and competes with) companies in regions that haven’t grown so fast. Thus, it can’t pass along all of its local cost increases—and labor is a local cost.

This phenomenon that labor is local and customers are not—which, of course, explains manufacturing in low-wage countries—is actually just one aspect of a larger reality. Manufacturers face various circumstances that affect their ability to compete for talent against other types of business.

A recent New York Times article by Adam Davidson captured the impact of this. Titled “Skills Don’t Pay the Bills,” the article observed that the apparent shortage of skilled manufacturing employees is not being accompanied by an increase in wages. Supply-and-demand says one should produce the other. Since the wage rise isn’t there, many question whether the “skills gap” is real.

It’s real. However, another peculiarity of manufacturing is in play here—the elasticity between demand and supply. Understanding this is valuable, as it explains much. Here is what I mean:

If a town has too few grocery stores, the market will know. This week, store shelves will be depleted. By next week, prices in those stores will go up, remaining up until a competing store opens nearby. 

Manufacturing is not like that. When manufacturers see demand beyond what they can comfortably accommodate, they are able to run uncomfortably for quite a while. Employee overtime is part of this; so is quoting later delivery dates. Manufacturers can also activate less efficient capacity, such as having employees run the shop’s older and seldom-used equipment in addition to the modern machines. In short, the manufacturing marketplace has all of this “stretch” to pull before prices and wages respond to demand.

One final factor in play is manufacturing’s cyclicality. CNC machining work is spotty even in good times—ask a job shop. The inconsistent demand makes it difficult for a manufacturer to predict how much farther its existing workforce and resources might yet have to stretch.

And that is where we are now. We’re stretching. Busy U.S. manufacturers perceive the need to attract and retain talent, but all of their competitors that are still managing to stretch limit their freedom to raise prices in order to offer more alluring wages.

How much more elasticity remains in the system? That’s a good question. Here is another: What happens to shop rates and wage rates—and what happens to reshoring—once this rubber band reaches its limit?