Unemployment Statistics Rely on Fuzzy Math

Fortunately, forecasting unemployment statistics isn’t a matter of life or death.   Because if it was we’d all be on life support. 

While we’d like to think that government reports on such highly anticipated and important economic indicators such labor statistics would be accurate within a few thousand people, it seems that the reliability of these mood-altering numbers are more like the odds we see in predicting Super Bowl winners than prescribing medicine.

According to an article in the Wall Street Journal, “it isn't all that simple to work out how many Americans are out of work. In fact, at a time when high unemployment is consuming many people,  the BLS can say only that it is 90% confident that the true change in the number of unemployed in March was somewhere between a drop of 243,000 and an increase of 511,000. In other words, it isn't even clear whether the number of unemployed rose or fell last month. The ranges are similarly broad for seven of the last 10 months—and for more than 75% of the time in the past decade.

It sounds to me that relying on monthly stats to guide us on important business and professional decision is slightly better than cracking open a fortune cookie.

There is no doubt that the unemployment rate remains near the highest it has been in decades. But the government doesn't really know how much that rate has been changing from month to month. That’s somewhat incredulous considering unemployment news causes stock market swings in the billions of dollars and significant shifts in consumer confidence.

Even the experts who are supposedly the captains of the ship navigating the turbulent economic recovery seas find the information unreliable.  "I make it my business to take all information with a grain of salt, given the inherent inaccuracies in the estimation process," says Richard DeKaser, who sits on the statistics committee of the National Association for Business Economics, a professional association of economists.

Why is there so much uncertainty in the unemployment number? Why can’t the government say whether more or fewer Americans are trying and failing to find work?

The jobless numbers are calculated by surveying a total of about 56,000 households in a small number of U.S. counties. At first glance, that number might seem ample. But the unemployment survey, conducted as a joint effort between the BLS and the Census Bureau, doesn't use a random geographical sample. That would require either costly and time-consuming in-person interviews across the country, or a telephone-only survey, which could risk providing unreliable information for a sensitive report.

These problems could be alleviated by increasing sample sizes. For instance, the survey on unemployment covers only about 17% more households than it did in 1967, even as the population has increased by about 55% in that time. But aside from the cost issues, visiting more homes yields diminishing returns—it would require a quadrupling of the sample size to cut that margin in half.

Sean Snaith, an economist at the University of Central Florida, sums up the flawed data, "You have to go with the data you have, not the data you wish you have."

So when next month’s U.S. unemployment rates are released, remember this –  eenie, meenie, minie, moe, the numbers releases can either be high…or low!

Link to original post

Avatar

Uncategorized

Leave a Reply

Unemployment Statistics Rely on Fuzzy Math

Fortunately, forecasting unemployment statistics isn’t a matter of life or death.   Because if it was we’d all be on life support. 

While we’d like to think that government reports on such highly anticipated and important economic indicators such labor statistics would be accurate within a few thousand people, it seems that the reliability of these mood-altering numbers are more like the odds we see in predicting Super Bowl winners than prescribing medicine.

According to an article in the Wall Street Journal, “it isn't all that simple to work out how many Americans are out of work. In fact, at a time when high unemployment is consuming many people,  the BLS can say only that it is 90% confident that the true change in the number of unemployed in March was somewhere between a drop of 243,000 and an increase of 511,000. In other words, it isn't even clear whether the number of unemployed rose or fell last month. The ranges are similarly broad for seven of the last 10 months—and for more than 75% of the time in the past decade.

It sounds to me that relying on monthly stats to guide us on important business and professional decision is slightly better than cracking open a fortune cookie.

There is no doubt that the unemployment rate remains near the highest it has been in decades. But the government doesn't really know how much that rate has been changing from month to month. That’s somewhat incredulous considering unemployment news causes stock market swings in the billions of dollars and significant shifts in consumer confidence.

Even the experts who are supposedly the captains of the ship navigating the turbulent economic recovery seas find the information unreliable.  "I make it my business to take all information with a grain of salt, given the inherent inaccuracies in the estimation process," says Richard DeKaser, who sits on the statistics committee of the National Association for Business Economics, a professional association of economists.

Why is there so much uncertainty in the unemployment number? Why can’t the government say whether more or fewer Americans are trying and failing to find work?

The jobless numbers are calculated by surveying a total of about 56,000 households in a small number of U.S. counties. At first glance, that number might seem ample. But the unemployment survey, conducted as a joint effort between the BLS and the Census Bureau, doesn't use a random geographical sample. That would require either costly and time-consuming in-person interviews across the country, or a telephone-only survey, which could risk providing unreliable information for a sensitive report.

These problems could be alleviated by increasing sample sizes. For instance, the survey on unemployment covers only about 17% more households than it did in 1967, even as the population has increased by about 55% in that time. But aside from the cost issues, visiting more homes yields diminishing returns—it would require a quadrupling of the sample size to cut that margin in half.

Sean Snaith, an economist at the University of Central Florida, sums up the flawed data, "You have to go with the data you have, not the data you wish you have."

So when next month’s U.S. unemployment rates are released, remember this –  eenie, meenie, minie, moe, the numbers releases can either be high…or low!


Link to original post

Avatar

Uncategorized

Leave a Reply