The flat line in the first graph is manufacturing share of GDP. The other graph of manufacturing output value shows it steadily increasing.
OK I guess I’m not smart enough to understand the difference between these two graphs, econbros @econophile @spidercrab help
https://www.stlouisfed.org/on-the-economy/2017/april/us-manufacturing-really-declining
I don’t know either. You can read this at the same time:
edit: I just got to the part where the author says the flat graph is wrong, but he doesn’t understand why.
I mean the first graph FEELS like what happened to the manufacturing sector in the US, right? Drive through Michigan in 1960 and 2020 and tell me different.
What about driving through Silicon Valley or somewhere in the South? That’s a lot of manufacturing isn’t it?
And jobs go down either way.
And say it’s somewhere in between…the real value of manufactured goods still increases even if not as fast as the growth of the rest of the economy. That’s not shrinking manufacturing.
If it happens for a few more decades at similar rates then it will be down to like 3% of GDP and we’ll just be making each other sandwiches as an economy.
The difference in the two charts is whether you adjust for inflation. There has been more inflation in non-manufacturing sectors than in manufacturing sectors. That’s why manufacturing’s share of nominal GDP (no adjustment for inflation) has decreased more than its share of real GDP.
OK. But if you asked an economist in 1960 what the manufacturing share of GDP was he’d say 26%, right?
I completely agree with the Justin Fox Bloomberg link that @microbet posted. I have no idea why you’d apply different inflation factors to sectors in order to arrive at some presumed “real share” measure over time. But, as Fox notes, it’s basically saying this:
Suppose all prices had remained at 1960 levels; what proportion of the total value of this year’s output would have been accounted for by manufacturing output?
That would mean that you’d take all the non-manufacturing output in current dollars and deflate it substantially more than you would deflate the manufacturing output in current dollars, increasing the apparent share of current day manufacturing. But, as Fox immediately goes on to note:
Why should we care about how the value of output would have grown had all prices remained at their year-b [1960] level? What’s so special about year b [1960]?
If you’re looking at industry share over time, it seems obvious to me that using nominal numbers each year is the right way to look at it.
I like how that post makes it seem like I was right. Let’s go with that.
Hate posting an article from Fox.