Replies

Using real data captures the increase.

So you look at the nominally adjusted data (all values adjusted to constant dollars i.e. year 2000 - so that your unit prices are all expressed in terms of year 2000 dollars) which shows a manufacturing decrease. Then multiply the technology sector by some large number (somebody's fudgefactor) because, your widget today is said to be functionally more capable the the widget of yesteryear. That is irrelevant when you are looking specifically at your level of taxable economic activity.

Perhaps whomever put this chart together used 1940 as a base year, adjusted all the dollars to a common base year, then asked what is the value of the capacity deployed in 2003 in terms of 1940 technology. The resulting number is then shown as the value of economic activity for 2003.

For instance: If in 1940 it would take ten typewriters and ten clerks one week to reproduce what you can do for yourself with a PC over the same period of time. They then compare the cost of the ten people and ten typewriters with the actual 2003 cost of a week of your time and your PC and come up with an adjustment. Something like, "the average office worker today is 8 times as productive as the office worker of 1940", and use that reasoning to either multiply 2003 amount spent on office workers by a factor of 8, or to divide the 1940 number by the same.

The result depends entirely on how you define equivalency between the products. If a car is a means to transport so many people over a given distance per gallon of gas, you get one answer. But, if a car is just so many pounds of steel you get an entirely different answer. You adjust whatever you want until you get the result that you are looking for. You are not adjusting everything, just those line items that you choose to modify.

This chart presents a distorted view which means nothing unless you are clear on how the numbers are being manipulated. We might be better off if we approach the question of US manufacturing from the perspective of global market share.

For instance: If in 1940 it would take ten typewriters and ten clerks one week to reproduce what you can do for yourself with a PC over the same period of time. They then compare the cost of the ten people and ten typewriters with the actual 2003 cost of a week of your time and your PC and come up with an adjustment. Something like, "the average office worker today is 8 times as productive as the office worker of 1940", and use that reasoning to either multiply 2003 amount spent on office workers by a factor of 8, or to divide the 1940 number by the same.

That's certainly a possibility, but maybe you want to use a different example. Your example involves a service. The chart involved goods.

From the chart link:
It is critical to adjust the figures for inflation in order to make a valid comparison. The price of many goods such as computers has fallen sharply. Since GDP data are calculated in money rather than volume terms, failure to account for price changes gives a distorted picture. For example, suppose the output of a product rose by 10 percent in terms of units while falling 10 percent in price due to higher productivity. Using nominal dollar figures makes it appear there was no increase in output. Using real data captures the increase.

See, units produced, not services provided which is how I read your office example.

You agree that there is a need to adjust when units increase while price drops, don't you?

Here's an example, how many beer cans were produced in 1940 vs in 2000? Do you think we produced more? No doubt. Do you think they were less expensive per unit? Again, no doubt. How would you adjust for that?