There are two ways producing widgets can grow 10%. The standard way is where production methods stay the same, and 10% more offices/factories are added and 10%

There are two ways producing widgets can grow 10%. The standard way is where production methods stay the same, and 10% more offices/factories are added and 10% more workers are added. The reason this would be done is because there was a 10% growth in demand for widgets. This would be due to a 10% growth in population that could afford widgets, or a 10% growth in income for consumers/workers that they decide to spend on widgets. With population growth shrinking in the standard consumer societies, that aspect goes out. Insofar as income, the average US inflation-adjusted hourly wage today is below what it was 43 years ago - it has shrunk. So the demand is not really increasing. Since demand for widgets in general is not going up, companies are not desperate to employ the average worker, and thus wages are stagnant (or as I said, the average US inflation-adjusted hourly wage has shrunk over the past 43 years). When were companies investing the capital necessary for automation, technological improvement etc.? From the 1940s to the 1960s. When demand was high due to increased population (baby boom). There was a baby boom because after a long depression, the economy started paying good wages to the average worker due to strong unions and government stimulus starting heavily during World War II. Money in workers/consumers pockets led to increased consumer demand. Due to the highest unionization rate in American history after World War II you have high wages at these growing companies. How to lower costs? Invest in automation, invest in technological improvement. Without high wages, and some of the other factors mentioned, the financial impetus to invest in technological improvement and automation fades. It's always there, but it ebbs in the absence of high demand, high wage employees.