On Friday, October 19th, Floridians will get another monthly indication of their state’s employment situation. However, if the national unemployment rate—which has already been released for September—gives any foresight into the Sunshine State’s, then the results may be too good to believe. Experts have heavily refuted the national figure, claiming that the Bureau of Labor Statistics’ (BLS) seasonal adjustment process needs to take into consideration the unusual times we live in.
National Unemployment Rate
The seasonal adjustment process helps to smooth out oscillations that tend to convolute the true story of our nation’s employment. In the case of the fall months, teachers resume their place in the work force and retailers begin to hire employees for the holiday season.
In September, the seasonally adjusted national unemployment rate dropped to 7.8%, a figure that hasn’t been rivaled since December 2008. Indeed, even the unadjusted figure fell to 7.6% this month; but given the relatively small survey size of the Current Population Survey (CPS) each month this statistic is in need of statistical “smoothing.” For example, the unadjusted national unemployment rate fell to 7.7% in April of this year, but the seasonally adjusted rate only fell to 8.1% in the same month. So why did the seasonally adjusted figure fall much further this month, and why are economists refuting the claim? The seasonal adjustment process helps to smooth out oscillations that tend to convolute the true story of our nation’s employment. In the case of the fall months, teachers resume their place in the work force and retailers begin to hire employees for the holiday season. This has been going on for decades, and the BLS uses (among other things) data from earlier years to help create the seasonal adjustment. There’s just one problem: during the great recession, unemployment began to snowball in the fall of 2008, which created a significant shift from the usual trend. Even though unemployment is still historically high, the U.S. employment situation has returned to its routine in terms of seasonality—that is, the unadjusted unemployment rate falls every fall (or autumn). Since the BLS’s seasonal adjustment software currently looks to 2008 for historical data, it tends to underestimate the seasonality of this decline.
Florida’s unemployment rate started 2008 at 4.8% and fell to 8.2% by year-end. This is because construction, prior to the housing bust, was a pivotal part of Florida’s economy, hence creating many jobs. Moreover, two other industries–which create a significant amount of jobs in Florida–are highly seasonal: education and retail.
Now turning our focus back to Florida, it seems likely that this effect will occur in this state as well—and perhaps even be somewhat magnified. In 2008, the national unemployment rate started the year at 5.0% and fell to 7.3% by year-end. Meanwhile, Florida’s unemployment rate started 2008 at 4.8% and fell to 8.2% by year-end. This is because construction, prior to the housing bust, was a pivotal part of Florida’s economy, hence creating many jobs. Moreover, two other industries–which create a significant amount of jobs in Florida–are highly seasonal: education and retail. Arguably, education is a pivotal part of every state’s employment structure, from elementary school all the way through post-secondary education. However, Florida’s retailers get an additional seasonal boost from “snowbirds,” or residents who only spend the fall and winter months in the Sunshine State. These factors, along with Florida’s overall employment trend, are depicted below:
Total Nonfarm Payrolls
In the second graph, the severe decline in construction jobs and the strong seasonal effect in government jobs (which include teachers) and the hospitality industry can be observed.
For retailers, the seasonal concern at the moment is whether sales will be robust during the holiday season, when many of them—toys stores, for example—will find out whether their years end in the black. The December spike is evident in the graph showing Florida’s gross sales (below). That spike is not as large as it looks, however. The vertical axis starts at $40 million, not zero, which visually exaggerates the amplitude of the spike. December gross sales average 42% above those in other months overall.
Florida's Gross Sales
Another type of seasonality in Florida has to do with summer and winter visitors. The southern part of the state specializes in winter visitors, the northern part in summer guests. The net result is very slight total seasonality (for retailers). Taxable sales average only 0.7% more in winter (January, February, and March) than in summer (June, July, and August). The counties with winter taxable sales more than 2% above summer are all in the south: Hendry, Collier, Indian River, Charlotte, Sarasota, Lee, Monroe, Highlands, Martin, Okeechobee, Glades, and Palm Beach. Those with summer sales exceeding winter are concentrated in the Panhandle, the large ones being Walton, Bay, Okaloosa, and Santa Rosa. Miami-Dade, Hillsborough, Orange, and Duval are balanced between summer and winter.
Overall, the winter months still end up with the lower unemployment, and hence need an upward seasonal adjustment. This adjustment seemed to be taken into account for Florida in August, when teachers went back to work—the unadjusted rate fell by 0.4 percentage points, but the seasonally adjusted rate held constant at 8.8%. However, the 2008 employment data may still produce similar results in Florida as it did in the nation, as Florida’s seasonally unadjusted rate went from 7.1% in August 2008 to 8.0% by year-end. Therefore, Floridian’s may have to keep a watchful eye on employment data going forward, cautious of increases late in the year.