Paul Krugman on "Economic Geography"

From The New York Times, the conclusion is this:

Take the (fairly celebrated) example of Rochester, New York. It started as a flour milling center, benefitting from the Erie Canal, then as a center for nurseries and seeds. So it was a resource-based center. Then, in 1853, John Jacob Bausch, a German immigrant, started a company making monocles, which became a major producer of glasses, microscopes, and all things lens related.

So Rochester became a place where people knew about optics, presumably creating the preconditions for the rise of Eastman Kodak, and much later Xerox. This was typical of small industrial cities: even if what a city was doing in, say, 1970 seemed very different from what it was doing in 1880, there was usually a sort of chain of external economies creating the conditions that allowed the city to take advantage of particular new technological and market opportunities when they arose.

Obviously, this was a chancy process. Some localized industries created fertile ground for new industries to replace them; others presumably became dead ends. And while a big, diversified city can afford a lot of dead ends, a smaller city can’t. Some small cities got lucky repeatedly, and grew big. Others didn’t; and when a city starts out fairly small and specialized, over a long period there will be a substantial chance that it will lose enough coin flips that it effectively loses any reason to exist.

That seems to be a fair description/explanation of how things are developing.

The article in turn links to lots of other articles, the one that triggered his article was one by Emily Badger, which describes/explains how agglomeration effects benefit the largest global cities but smaller regional cities are losing out; also the links between global cities appear to be stronger than the links between any global city and its own hinterland/host country.

UPDATE: Now I've thought about it, there's a similar phenomenon with car manufacturing nowadays, which I wrote about here, it's just that large car factories are not concentrated near global cities, they are wherever they are for historical reasons. Most of the hundreds of small factories there used to be have closed down and a few have swollen enormously.

While this is all good stuff, what puzzles me is why anybody thinks these are blinding new insights? They aren't. I'm a land value taxer because I look at the real world and draw real conclusions. I know which factors drive (or depress) land values, and those factors are pretty much the same as those which encourage the growth of global cities and cause the gradual decline of smaller regional cities, in a word, 'agglomeration'.

Replacing taxes on earnings and output with Land Value Tax would ameliorate the effects of all this; it would drive growth in the largest cities; reduce the drag of taxation on smaller cities and 'the regions' (where land values are lowest); and act to redistribute the benefits of economic growth more evenly across the country - an overall levelling upwards.