Sunday, March 6, 2011

Unemployment vs. Trade

Click to enlarge.

Here's a simple explanation of the x-axis. A country that only exports would be seen as 100% in the chart. A country that only imports would be seen as -100% in the chart. A country with balanced trade would be seen as 0% in the chart.

Note that there is a lot of noise in the data, but the trends do seem valid to me. The higher the trade deficit, the higher the unemployment. I've shown two trend lines. The trend line in red weights all countries equally and the trend line in blue weights all countries by their GDP. The slopes of both trends are consistent.

Also note that a 10% unemployment rate in the United States is entirely consistent with the global weighted linear regression trend line based on our current trade situation. I would argue that unsustainable government deficit spending is the only thing keeping it below that rate.

I listed all the countries from the CIA World Factbook that have data for GDP, unemployment, population, exports, and imports. If any country did not have all of that data then it was excluded from the list. For example, Somalia was removed. Perhaps they would simply prefer to pirate.


Since 1991 no central government has controlled the entirety of the country, despite several attempts to establish a unified central government.

Piracy in Somalia

There have been both positive and negative effects of the pirates' economic success. Local residents have complained that the presence of so many armed men makes them feel insecure, and that their free spending ways cause wild fluctuations in the local exchange rate. Others fault them for excessive consumption of alcoholic beverages and khat.[26]

On the other hand, many other residents appreciate the rejuvenating effect that the pirates' on-shore spending and re-stocking has had on their impoverished towns, a presence which has often provided jobs and opportunity when there were none. Entire hamlets have in the process been transformed into veritable boomtowns, with local shop owners and other residents using their gains to purchase items such as generators -- "allowing full days of electricity, once an unimaginable luxury."

Speaking of pirating, the Cayman Islands sure are bucking the trend (as seen in the chart). How do they do it?

September 14, 2006
Offshore tax shelters not just for the rich

• Cayman Islands. United Kingdom territory is home to more than 500 banks and trust companies, 7,100 mutual and hedge funds.

And then there is Turkmenistan. What a joy it must be to live there. I would not have even thought to check out what is going on in that country had I not made this chart.

February 28, 2011
No student protests here, they wouldn’t dare – welcome to Turkmenistan

These measures add to the already restrictive atmosphere in which academia operates. Students at Turkmen colleges and universities are not allowed to appear in foreign media, leave the country on vacation, drive a car, or use mobile phones on university premises. There is of course, little chance of protest within Turkmenistan. This is another right that Turkmen citizens do not enjoy.

Source Data:
CIA - The World Factbook
Weighted Linear Regression Instructions


Stagflationary Mark said...

How Long Until We Reach Full Employment?: Using the Last Two Recessions as Guides

Here's a better question in my opinion.

How Long Until We Reach Full Employment?: Using the Last Two Bubbles as Guides

If we're going to look for answers in the rear view mirror then we might just as well look closely.

Who thinks a commodity bubble will create 4% unemployment? Anyone? I sure as heck don't.

Stagflationary Mark said...

The Hollowing Out of the U.S. Income Distribution Under the Pressure of Technology

I don't see a problem with the number of jobs: I don't see any reason that technological unemployment should be any more in our future than it has been in our past.

I do and I think many will be blindsided by it. Our last two bubbles covered up the details. This next bubble will put the nail in the coffin.

Just an opinion of course.

AllanF said...

If it wouldn't be too busy, can you highlight the dots for China & Japan on that chart?

As for technology destroying jobs, the problem all the ivory tower economist bozos and fascist elected officials seem unable to grasp is technology is supposed to destroy jobs, that's the whole point. The problem is, without unions or regulations, the capitalists will keep all the productivity gains for themselves. They can't help it, competition is cut-throat and they can't unilaterally disarm even if they wanted to. But what the friggen' idiots could have done is pass laws and regulations that forces the capitalists to share the wealth such as: decouple health insurance and full-time employment; reduce the work week; promote job sharing; stop importing foreign labor that both undercuts the native wages and makes up the difference by burdening government social services; apply heavy marginal tax rates on the 'lucky' winner-take-all tech monopolists; stop trashing the dollar in a race to the bottom in labor costs.

Stagflationary Mark said...


I hear that!

I don't have access to my computer (I'm on the PS3) but I can help you find them.

China's Trade # (x-axis): 7%
China's Unemployment: 4.3%

Japan's Trade # (x-axis): 9%
Japan's Unemployment: 5.1%

Stagflationary Mark said...

Since our trade # is -20% and China's is 7%, a rational finger pointer might argue that we are 3x the trade imbalance enabler that China is. Sigh.

Mr Slippery said...

Re: Turkmenistan

First, any country ending in -stan is likely be a rockin' vacation resort. Turkmen can surf the Caspian, ski in nearby Iran, and gamble the night away at one of the famous Uzbek casinos. I think Mandalay just opened a new property there. And don't forget the perfect weather.

Let me hear you balalaika's ringing out!

AllanF said...

Thanks Mark. Now that I've gotten a little sleep I think the most interesting thing is how the two regressions are so similar in slope.

Unless, I suppose, the non-developed countries are so random that they cancel each other out and the regression simply reverts to the weighted GDP + an unemployment offset. I reckon the R-squares could speak to that.

Stagflationary Mark said...

Mr Slippery,

First, any country ending in -stan is likely be a rockin' vacation resort.

Indeed! ;)

Stagflationary Mark said...


When I saw the slopes match so closely my first thought was that I had made a mistake in the calculations.

I therefore bumped the Cayman Island unemployment rate up to 4,000% (not that it could really go that high) to see what would happen. The slopes no longer were even close and did what I would have expected, so I think I did it right.

The r-squared values are incredibly low since there is so much noise in the data. There are clearly much bigger factors than trade that affect the unemployment rate.

The most interesting thing in the chart to me is...

It isn't hard to find a country that will import virtually everything (compared to exports) but there aren't any countries that will export virtually everything (compared to imports).

Free lunches are easier to swallow perhaps. ;)

Jazzbumpa said...

I would argue that unsustainable government deficit spending is the only thing keeping it below that rate.

I very seriously doubt it. I was just looking at spending data today when my lovely new lap top blew up and took all my data files with it.

Spending growth is not out of line from any historical trend. I wanted to get that on a log scale to demonstrate. Maybe some day soon.

Deficits are mainly due to revenue short falls - either from unwise tax cuts, as we've seen for most of the last 39 years, or - as now - from a depressed economy reducing the tax base.

On the spending side, only one thing has ever made a big difference, and that's spending on the military.


Stagflationary Mark said...


"Spending growth is not out of line from any historical trend."

"Deficits are mainly due to revenue short falls..."

If I am living paycheck to paycheck, my employer cuts my pay by 20%, and I continue to spend the same amount each month then I am using unsustainable deficit spending to make up the difference.

You could argue that the revenues will rise and this effect is therefore only temporary and cyclical. I'm in the structural camp though. The revenue increases from the dotcom and housing bubbles were temporary. We should not count on them returning.

Stagflationary Mark said...

Federal Government Current Receipts (FGRECPT)

We've had two bubbles so far. In my opinion, we're working on #3.

Stagflationary Mark said...

I intend to adjust that last link for population growth and inflation. It will turn even uglier when I do that.

Thanks for the inspiration. It will be the topic of my next post.

nilys said...

I wonder whether any of these economic indicators – GDP, U3 unemployment, CPI, etc – still have any meaning. It may be that economic reality somehow transcended the old parameters set up to capture it and we need new indicators.

I don't know whether your graph is any better than accidental. Each country has its own way to measure unemployment (even through each may subscribe to the same definition of who the unemployed are). We have the U series. I just looked up what definition of unemployment China uses and found this ( ). Glancing through the paper, seems like officially only those who are registered are counted as unemployed. Again, do the unemployed have an incentive to register or not? If not, then obviously this indicator is a low estimate. Furthermore, the officials can potentially manipulate in subtitle ways the incentives for the unemployed to register. Cultural predispositions and attitudes may also control whether the unemployed register or not. Black markets may exist of varying size. The authors believe the official number in China is an underestimate and go on to discuss alternatives. The paper is a bit old, but then, as is now, the official number was around 4%.

I also found this titbit:
“Persons without formal authorisation to live and work outside rural areas and who have moved to the cities to look for work cannot register at labour exchanges and are therefor excluded from the unemployment figures. Similarly, all workers laid off from state enterprises that since 1998 have been entitled to social welfare benefits also cannot register as unemployed”.

As for technology, I believe that it is both a creator and a destroyer of jobs. Sure, increases in efficiency due to technological innovation reduce workforce. But at the very same time technological advances open up huge new employment fields. A most obvious example of a new field: personal computing and the internet. A lot of it has to do with human imagination, changing preferences and willingness to adopt new technology and to embrace new ways of life. The present period in the US seems to me to be a low point in spirit and in imagination, a period of clinging on to old no-longer sustainable and even unhealthy modes of living and of refusal to move on and reinvent ourselves.

Jazzbumpa said...

Mark -

Going back to the tax rates that were in effect when times were good would take us back to good times.

Our current problems result from the worst wealth disparity ever, and the resulting misallocation of resources.

There are three (not necessarily a mutually exclusive and possibly sequential) outcomes. The next great depression, So. American style banana republicanism, and blood in the streets.

The thing that's unsustainable is wealth disparity.

The vast majority of the problems are on the revenue side.


Stagflationary Mark said...


I don't know whether your graph is any better than accidental. Each country has its own way to measure unemployment (even through each may subscribe to the same definition of who the unemployed are).

Could be. One would think that the differences in measurements would tend average out a bit but one never really knows.

Thanks for the comment. Blogger marked it as spam. It clearly isn't!

Stagflationary Mark said...


The thing that's unsustainable is wealth disparity.

I do think imbalances are the root of the problem and we have so many imbalances to choose from. Sigh.