How the National Institute of Statistics and Economic Studies (INSEE) conceals French disaster about creating companies and jobs

Focus on a deliberate misinformation.

If an economist visits Eurostat statistics, he will find 407 051 new German jobs created in 2007 for 395 053 new French jobs created in the meantime. Nothing to worry about, especially given that Germans are about 80 million compared to slightly more than 60 million French citizens.

The truth is nonetheless that we are creating 200 000 fewer jobs per year than the Germans.

To measure job creation, the Census Bureau (the American equivalent of INSEE) only takes into account the workforce in companies with at least one employee. Europe should follow, as evaluating the real working population in firms without employees is pie in the sky.

How did INSEE find those additional nearly 400 000 created jobs? We have analyzed the 2002 figures, similar to those of 2007 but with detailed distribution.

First, INSEE counted as newly created jobs the creation of companies without employees, even when the company is an empty shell. Following the American model -which may be followed by Eurostat in the upcoming years, we only took into account jobs created in companies with at least one employee.

Secondly, INSEE counted as newly created jobs the ones that previously existed but have been purchased or reactivated.

This allows to straighten the differences

These figures published by INSEE in 2002 are consistent with the 104 500 jobs created in the more recent 2004-2008 period and published by the APCE (a business start up agency) with INSEE figures and used by firm-demography.

INSEE’s overrating had already been noticed by the OECD in 2006: “The proportion of start-ups which are not real creations varies depending on sources […]. In Europe, this proportion is usually around 20%, while French national figures suggest a 30%-40% proportion.”

Without doubts, IRDEME figures underestimate the real number of newly created jobs. To avoid hiring, or because social security contributions are lower for an independent than for an employee, lots of companies without employee are real job creations and are not “empty”.

Nonetheless our conclusion remains the same: as soon as we consider companies with more than 2 employees, we create 2 to 3 times fewer jobs than Germany or England. (See chart below).

If we want to create companies such as Google, Intel or Federal Express, it is unlikely that they are born without employees: a real project implies hiring several people from the very first day, to enable large-scale and fast innovations.

The gap between French and German companies is visible:

This chart shows the Pareto distribution of newly created jobs in France and Germany, for a comparable population. It gives for each size (in number of created jobs) the number of companies created with at least this size. For example, the gap between both charts at the level “more than 10 employees” shows that Germany creates more companies of this size by 50%.

INSEE’s figures of job creation in France may respect Eurostat rules. But given their crucial role, it is deliberate misinformation. Is INSEE’s role only about publishing data? Or should the Institute alert our country about the economic dangers which threaten its survival?

* "employer companies"