Global youth unemployment: Making sense of the numbers
High youth unemployment is one of the biggest problems confronting societies around the world - but do we really know how bad the situation is?
The statistics are terrifying - the United Nations' International Labour Organization (ILO) estimates that close to 75 million 15 to 24-year-olds around the world are out of work. But how accurate is this number?
While the long-term costs of being young and jobless are not in doubt, we cannot say the same about the figures which are published in this area, because of the way unemployment rates are calculated.
All unemployment rates - including youth unemployment - are calculated as percentages not of the total population, but of something called the "economically active population".
Young and jobless in numbers
of young people are
more likely to be jobless
are not in education or training
Youth unemployment is highest in
North Africa - 27.9%
It is defined as the employed plus the unemployed - but that leaves a huge number of other people out of the calculation.
This is because most countries calculate their unemployment rates by carrying out regular labour force surveys which ask people a series of questions which help place them into one of three categories.
But even these definitions might not be what you would necessarily think:
- Employed: Someone who has performed any work. Even if it is just one hour a week, someone is counted as employed if they have worked for "family gain". So, someone who is delivering newspapers for a couple of hours a week is classed as being a worker. So is someone who helps out unpaid in a family business - people do not have to have been paid in order to be counted.
- Unemployed: A person is classified unemployed if they do not have a job, but would like one, have actively looked for one and have the time to do it.
- Economically inactive: Someone who is neither employed nor unemployed. So, it could be that the person does not want or need to work, so has not looked for a job. Or it could be that the person cannot work for a specific reason, such as bad health or studying full-time (although some students may have part-time jobs, and so be considered employed).
Distorted unemployment figures
The three classifications listed above are suggested by the ILO - but when they are applied to young people it can lead to some quite strange results.
The unemployment rate is calculated by dividing the number of unemployed people by the number of economically active people, which includes both employed and unemployed people.
The more young people who choose to stay in education, the bigger the pool of economically inactive young people grows - while the pool of young people who are economically active shrinks.
Any increase in the proportion of young people who decide to delay entering the labour market and keep studying - and remain economically inactive - has the effect of increasing unemployment figures because the number by which the unemployed population gets divided by shrinks.
But importantly, the overall number of young people stays the same.
The slider shows how it works. Slide the arrows.
As the number of people classed as inactive increases (2,560,000) so does the youth unemployment rate (21.6%). But the number of unemployed people stays the same at 1,020,000.
*Inactive can include students. The data is from the UK's Office for National Statistics, July 2012.
So, as more and more young adults choose to stay in education, the economically active population - the denominator used to calculate the unemployment rate - reduces, and appears to drive up unemployment.
As a result youth unemployment rates can distort the picture of how bad the job prospects are for a country's young people.
"I think it's important to advise against the use of any single indicator," says Ekkehard Ernst, the ILO's head of unemployment trends.
"You cannot only take the unemployment rate and say this is my only measure of problems in the labour market.
"This is because - even for the adult population - you have differences in the possibility that people have to draw from unemployment benefits.
"You also have differences in the tax systems that incentivise people to participate or not in the labour market."
In the developing world, where it is not so easy to stay in full-time education and there is not the same level of unemployment benefits to fall back on, young people are often forced to take any work that is available, which can be poorly paid and sporadic.
So, the rate alone often does not reflect the true problems of people finding appropriate jobs with a regular wage attached.
For these reasons some argue that youth unemployment is better and more accurately represented by using a ratio, calculated as the share of unemployed, as a percentage of the whole youth population.
And if you use these, things look rather different.
Take Spain - the country with the worst youth unemployment rate in Europe in 2011 at 46.4%.
If the ratio is used instead, the figure is more than halved to 19% - in troubled Greece it is even more pronounced.
While Greece had an unenviable youth unemployment rate of 2011 of 44.4%, its corresponding ratio was 13%. (Source: Eurostat)
These are still very high numbers with real people lying behind the statistics, but it paints perhaps not such a stark picture.
Another major problem with global youth unemployment statistics is the quality of the data itself.
Data collection methods vary between countries - many use different age limits to define exactly who is classed a young person. While in others the data can be old or missing altogether.
When unemployed is not unemployed
Some countries do not even count first-time unemployed as being unemployed at all - clearly a problem when applying that idea to young people.
These difficulties can be seen in the data from the two regions highlighted by ILO research as having the highest youth unemployment rates - the Middle East and North Africa.
In these areas, the situation looks bleak. Youth unemployment in the Middle East is 26.5%, with 3.4 million young people without a job. In North Africa the rate is even higher at 27.9% - 3.85 million jobless.
The official figures for countries in these two regions are wildly out of date, with some data missing entirely.
Four countries make up the North Africa region - Algeria, Egypt, Morocco and Tunisia. But the most recent unemployment numbers for Egypt were published in 2007. Tunisia's most recent data is from 2005.
Both of these countries have seen enormous upheaval since this data was published, as a result of the Arab Spring.
If youth unemployment predictions were based solely on this research, the total number of young unemployed in the Middle East is 796,000 - a quarter of the ILO's estimate - and 2.4 million in North Africa.
Measuring youth unemployment in sub-Saharan Africa is even more tricky, says the ILO, where much of the data is between 10 to 15 years old - or missing completely.
Given the figures it has for the rest of the world, the ILO says its global figures are "relatively reliable" but because of the limitations caused by countries with old or missing data, the ILO relies on econometric models to fill in the gaps.
"We are regularly updating the models that we use for countries. We do take into account economic and political circumstances in order to make sure this is reflected," Ekkehard Ernst explains.
"Our unemployment estimations for North Africa has gone up dramatically since 2011 and we're projecting further increases over the next year.
"This is as best and as close as you can get to the reality if there is no information truly available at the country level."
Unemployment is undoubtedly a huge obstacle confronting millions of young people across the globe.
But as even the ILO concedes, there is huge difficulty in comparing national data, which is often inconsistent, incomplete or simply unavailable.
This problem is compounded by the distorting effect of large numbers of young people remaining in education and so a degree of caution should be exercised when quoting unemployment rates.
"There is an element of uncertainty in there," says Ekkehard Ernst.
"We do have an idea of this indicator being relatively precise for developed economies because it's directly measured and these are typically the countries where most of the measured unemployment is happening.
"But if you look at the global figure then it is true that you have to take it with a grain of salt."