With a sociopathic insouciance that I strongly approve of, Matt Yglesias suggests that the essence of manufacturing (vs services) is putting stuff in boxes and that’s not really something anyone should care about.
Kevin Drum disagrees
There really are some good reasons to care about manufacturing jobs. Here are three:
- The manufacturing sector is generally more capital intensive than the service sector. Because of this, a pea canning factory can afford to pay higher wages for unskilled and semi-skilled labor than a restaurant can.
- On a related note, manufacturing facilities are generally more scalable and more amenable to technological improvements. This improves productivity, and improved productivity is key to improved wages. By contrast, the restaurant business doesn’t have a lot of scope for automation or productivity improvements.
- Manufacturing is part of the tradable sector, while service industries generally aren’t (though there are exceptions). A pea canning factory can ship its products overseas and help maintain our balance of payments. A restaurant can’t.
So, none of these things are necessarily true.
Sweatshops tend to be highly labor intensive. A radiologist’s office is highly capital intensive.
The West Edmonton Mall is about 3.8 million square feet. That’s about the size of GM’s largest facility, Arlington Assembly. Though, admittedly both are dwarfed by River Rouge, which at its peak claimed 16 million square feet of floor space and was a thing of beauty:
Still, the Abarj Al Bait Towers in Mecca claims 16.7 million in floor space. Though, it contains residential as well as shopping and hotel, so its not really pure “service” facility in the same sense.
Virtually all services are tradable. You just need to move the customers rather than moving the goods. When Euro hit highs just before the crisis New York stores experienced an influx of European shoppers.
What is important, however, is this: putting things in boxes raises the catchment area for most facilities. The fact that there is a box involved means that I can service customers all over the world. That in turn means that a manufacturing facility can benefit from the type of agglomeration effects that services often depend on cities to achieve.
This is also why hotels tend to be the largest and most capital intensive consumer service facilities. They are catching tourists from all over the world. Clearly, that’s also why it makes sense that the very largest of them all is in Mecca.
The decline of manufacturing is the fundamental reason why some economists predicted the rise of the city. To achieve the same scale affects people would have to be in close proximity to one another.
In the near term they seem to be right, though I still hold on to the notion that cheap energy and telecommunications will reverse the trend and allow services to be provided from anywhere.
In any case the key question, as it has been for hundreds of years, is the extent of the market and how much specialization and capital intensity that allows. Putting things in boxes gives manufacturing an advantage but not an insurmountable one.

9 comments
Comments feed for this article
Tuesday ~ December 13th, 2011 at 1:47 pm
Joshua Probert
It’s also important to note manufacturing is becoming less capital intensive.
3-D printing, sending parts of for rapid prototyping, manufacturing that is customizable (esp. electronics manufacturing) means you can sell the excess capacity of those machines that *are* capital intensive.
If you consider app/software/web development manufacturing this is also extremely low cost to start up.
Amazon gives you a free year for their lite-sized instances.
Tuesday ~ December 13th, 2011 at 1:49 pm
Joe Carter
With a sociopathic insouciance
At what point will real economists stand up and admit that Slate’s “business and economics” columnist knows nothing about business and economics? Since he was nothing but a kid he’s been writing (mostly about stuff he has no experience with) as if he knows what he’s talking about. And people who know better just play along. When will it end?
Tuesday ~ December 13th, 2011 at 4:27 pm
ed
I’m a real economist, and it’s pretty obvious to me that Yglesias knows and understands lots about economics. Certainly much more than most journalists, and even more than some Ph.D.s I’m aware of.
Tuesday ~ December 13th, 2011 at 2:24 pm
Nick Rowe
Even if you assume manufacturing is more capital intensive than services, is that a good thing for wages? If you wanted to increase real wages, would you subsidise a capital intensive or a labour intensive industry?
Tuesday ~ December 13th, 2011 at 2:24 pm
Alan Robinson
Another reason to focus on manufacutring is that two major sectors of the economy, retaila and construction are going through major transformations. Retail sales are shifting to the web. Web based sales require significantly fewer employees than brick and mortar sales and the jobs are usually found in cities far from where the customer lives as opposed to the same town.
Construction is in a slump, not just because of the overcapacity of residential real estate but as a side effect of the switch in retail spending to the web. The overcapacity in retail real estate is now causing malls to be torn down if there is another use for the land. In cases where there is not, like in Erie PA, they lie vacant in decay. The problem of commercial construction may be almost as bad as office space shrinks as space that used to be use store paper are gone.
Tuesday ~ December 13th, 2011 at 2:56 pm
Bare-faced paternalistic contempt « Blunt Object
[...] of technocracy (not to mention the examination of emotionally-freighted economic issues with “sociopathic insouciance“). With that in mind, I’m trying to take my radical Hayekian/Rothbardian [...]
Tuesday ~ December 13th, 2011 at 3:47 pm
Wonks Anonymous
“Virtually all services are tradable. You just need to move the customers rather than moving the goods. When Euro hit highs just before the crisis New York stores experienced an influx of European shoppers.”
It’s a lot easier to move manufactured products than customers.
Tuesday ~ December 13th, 2011 at 6:11 pm
Lord
Seems quite insurmountable to me. Manufacturing is worldwide from the factory door. It allows for the greatest specialization, specialization in machines making machines, the robot being the end product. And the greatest in reproducibility and scalability which services are barely at all and not at all without more labor inputs. This means that other than protected products, patents and copyrights, income differentials between workers will be limited and opportunities for wealth constrained (perhaps not a bad thing).
Wednesday ~ December 14th, 2011 at 5:25 am
Axel
To me your radiologist example does not really answer to the first argument:
“The manufacturing sector is generally more capital intensive than the service sector. Because of this, a pea canning factory can afford to pay higher wages for unskilled and semi-skilled labor than a restaurant can.”
This statement is about employment and wages of UNSKILLED workers. This is clearly one big challenge for large developped economy, which even with huge education efforts can’t really dream of having only skilled workers in their economy anytime soon.
Besides, the technological production system has changed with IT at the end of 20th century. 19th century 2nd indsutrial revolution based on mechanics, oil and chemicals is one of the past. Yet, this ‘industrial revolution era’ had one great social advantage which was to allow fordism and middle class development via giving a real profitable production role to un(low)skilled people. Now we are entering a software/IT era where innovation & skills are more or less the only key advantage for a company to survive. This means unskilled labor, while globally more abundant than ever, will suffer rom huge & consistent pressure on real wages for long (and developped/wealthy world will keep high unemployment as you can’t live with a Chinese wage in the US for now). I doubt services can compensate for this expensive unskilled labor supply in developed world without some loss in potential GDP.