# Has the natural rate of unemployment decreased with modern travel and the Internet?

The natural rate of unemployment is a combination of structural unemployment and frictional unemployment.
Structural unemployment is caused by a surplus of labor at a given wage in a given area.
Frictional unemployment is caused by workers searching for or transitioning between jobs.
However, modern travel (airplanes, automobiles, etc.) would reduce structural unemployment and frictional unemployment by allowing for faster movement of excess labor to transfer to areas with a shortage of labor; similarly, the Internet (specifically search engines, online jobs fairs, etc.) would reduce frictional unemployment by allowing for workers to search for and find jobs more easily.

Does this imply that the natural rate of employment decreased with modern travel and the Internet?

To me, it seems that it has increased, not decreased, due to the factors you mention. Yes, transportation and information networks enable workforce movement. But they also enable movement of goods and information - and because goods and information are more mobile than humans, they profit more, and the results of their portability outpaces the results of the workforce's increased mobility.

When goods were hard to move, a local demand had to be met by local suppliers. If a manufacturer lost a deal, it went to a local competitor instead, and the workforce could adapt easily. Now, if a manufacturer loses a deal, it goes to a competitor 5000 km away. Take for example the Ruhr area: it's still the most densely populated area in Germany, even though the steel industry which attracted workforce in the mid-20th century is gone. The Germans literally shipped one of their steel mills around the globe to China - but the workers stayed.

So, because the workforce cannot adapt to structural changes with the speed at which they are happening, the structural unemployment is higher than before networks which enabled such changes. But the increased speed of change also means more frictional unemployment. A large part of the adaptation which is happening requires people to change jobs. Also, your observation of an increased commute radius contributes to frictional unemployment too - when a person is unhappy with his job, he has more opportunities in his now increased commute area, and is more likely to change jobs.

There are lots of indirect effects too. The Internet and the transport networks enable changes in culture (e.g. today people move out of their birth town early, so may be more willing to relocate again later), business models (e.g. movie streaming) and market structure (today's trend towards ever larger corporations is enabled through, among other things, efficient information and people movement, which reduces overhead costs of giant companies, and global knowledge - I doubt that 50 years ago, an American chain for mediocre coffee in fancy large cups would have been able to establish itself in Europe. Nowadays, the culture export means that the trendy European teenager recognized the brand before they had one of the shops in their location). And they go both ways - a giant corporation may actually be more stable, and afford to keep workforce during short-term demand slumps which would have capsized a tiny manufacturer.

Whatever globalization (spearheaded by more exchange of information and goods, and even people on their respective networks) may bring in the future, it currently seems to create more unemployment - not because it's a bad thing in itself, but because in a more dynamic market, there is less job stability.

• Do you have any references or data for your main claim, i.e. that information networks increased structural unemployment? – FooBar Apr 29 '15 at 10:43
• @FooBar no, I have no references. The answer is based on my impression of the current state of the economy, which is constructed from many bits snapped here and there - news, observing friends who are seeking and/or changing jobs and hearing their arguments for/against commuting, and so on. And from there on, it is an analysis of the situation done as best as I can. If you have references which show flaws in my reasoning, or data contradicting my observations, I'd be very interested in seeing it. – rumtscho May 13 '15 at 12:16
• I see a flaw here: because goods and information are more mobile than humans, they profit more, and the results of their portability outpaces the results of the workforce's increased mobility. I can agree on "they profit more". That is, potentially, increased travel and technology has benefitted capital more than labor. This, however, is not equal as saying that labor is actively harmed by the outcome. – FooBar May 13 '15 at 12:44

I expect that in the UK it has increased due to the benefit system, we now have lots of unemployment in some parts of the north, but little movement of unskilled labour.

However when the mills were built worker moved from all over the UK to them.

Therefore I don’t know how the different effect can be separated.

This was supposed to be a comment to @rumtscho, but it turns out to be too long, and then I kept adding information, so perhaps this deserves to be its own answer

Is an increase in mobility really detrimental for efficiency? Imagine a world with non-mobile capital and labor. Now we increase mobility of capital (but not labor). Now @rumtscho argues, this would decrease employment. In general, the opposite should happen: A decrease in rigidity increases efficiency (in almost any model I am aware of), and hence the utilization of previously unused resources.

Intuition Intuitively, imagine some people being unemployed in region XYZ, because they are stuck in a region with labor but no capital. With mobility of capital, capital will move towards that region (because unused labor promises higher returns to capital).

### General Case

Technically, the change over time of the unemployment rate $u(t)$ can be denoted as

$$\dot u(t) = s(t)(1-u(t)) + u(t)e(t)$$

where $s(t)$ is the rate of separations, and $e(t)$ is the rate at which unemployed find employment.

Now, technology can affect both $s(t)$ and $e(t)$. I have brought forward reasons why $e(t)$ may increase, and @rumtscho has brought forward reasons for why $s(t)$ can increase.

Conclusion The overall outcome is unknown. Anecdotal evidence and intuitive hand-waiving arguments have betrayed the economic profession quite often. To answer the question at hands, we need real data. Of which I, unfortunately, am not aware of.