Op-Ed: Torturing employment data
Dr. John Stewart, who worked for many years as an economic advisor at the Puerto Rico Industrial Development Company, used to say that if you tortured the statistics long enough, they would tell you what you wanted to hear. Lately there have been plenty of tortured statistics in Puerto Rico.
The Labor Department regularly publishes two employment surveys, the establishment survey and the household survey. The former is useful, but the latter is not.
The establishment survey uses statistics provided by employers who pay the fees for unemployment and disability benefits. These employers include only those who submit W-2 forms and pay Social Security. However, because the establishment survey doesn’t interview individuals, it offers figures for the number of people who are employed, but not for those who are unemployed.
The establishment survey is a good indicator of the economy’s performance. So much so, that the Government Development Bank uses the establishment survey as input for its Economic Activity Index (EAI). This Index has a fairly good acceptance among the community of economists on the island.
Meanwhile, the household survey asks individuals if they are working for a company, working for themselves, unemployed and looking for a job, or unemployed and not looking for a job. This survey has been the subject of criticism and skepticism because it has been inconsistent with other economic variables. As a result, it is useless to gauge the performance of the local economy.
For example, in August three friends start the month unemployed. One decides to sell “piraguas” (snowcones) and another one returns to college. The third decides to emigrate after her partner loses his job that month. In this scenario, the establishment survey would show a loss of one job because of the partner who has just lost his job.
In contrast, the household survey would show employment as constant because one person (the partner) lost his job, but another started working for herself selling “piraguas.” Moreover, unemployment would show a decline because the three friends are technically no longer looking for a job in Puerto Rico.
In August 2012 the establishment survey showed a loss of some 6,000 jobs compared to August 2011, consistent with an economy that stopped contracting but is still stagnant. Meanwhile, the household survey showed an increase of 20,000 jobs over the same period, consistent with an economy growing strongly.
These are the statistics. Which survey reflects the true performance of Puerto Rico’s economy? We invite the reader to draw their own conclusions without torturing them.