By Christine Smith, Public Affairs Staff
"Residual seasonality” sounds like the lingering impact of a long, cold winter stuck indoors: Sorry, I’m not ready for the beach. Residual seasonality has made my pants tight.
The real definition matters to monetary policymakers and others trying to make timely assessments of the nation’s economic performance by looking at gross domestic product, or GDP.
You probably know that GDP is a popular measure of the size of the U.S. economy and how it’s faring from one period to the next. A 2018 Page One Economics lesson explains that GDP measures the total market value, expressed in dollars, of all final goods and services produced in our economy.
Why is GDP a useful indicator? “When compared with previous periods, GDP tells whether an economy is producing more output (expanding) or less output (contracting),” writes author Scott Wolla, St. Louis Fed senior economic education specialist.
A ton of information gets rolled up into GDP. For example, if you buy a domestically produced car or a carton of milk, that’s in GDP. If you pay for a haircut, that’s in GDP. If a private business or the government buys new machinery, that’s in GDP. Data are collected by the Bureau of Economic Analysis, which estimates the nation's GDP for each year and each quarter.
Think of how consumer patterns tend to vary throughout the year—back-to-school, winter holidays, etc. When experts examine economic data, they account for “seasonality”: predictable ups and downs that occur due to seasonal events.
As St. Louis Fed Economist Michael Owyang and Senior Research Associate Hannah Shell explain in a recent Regional Economist article, such predictable variations in economic data can be caused by:
Economists usually remove seasonality to compare data, such as GDP, across consecutive quarters. That’s known as deseasonalizing.
But Owyang and Shell note that during the past few decades, GDP growth in the first quarter has been substantially lower than growth in other quarters—even after adjusting for typical seasonality. The figure below shows this pattern of weaker GDP growth. It looks at three periods: 2005-09, 2010-14 and 2015-18. In each, the authors note, “average GDP growth is markedly lower in the first quarter.”
Owyang and Shell explain, “Economists have dubbed this phenomenon ‘residual seasonality,’ suggesting that the published first-quarter GDP growth rate is artificially low and that actual growth is more in line with that of the other quarters.” Put another way: “[R]esidual seasonality is the leftover seasonality that remains in already deseasonalized data.” (Hence the title of their article: Dealing with the Leftovers: Residual Seasonality in GDP.)
So, residual seasonality is kind of like finding stray tinsel under the couch after you cleaned up the holiday decorations. You thought you had deseasonalized the house.
Owyang and Shell dig into why leftover seasonality might be hanging out in seasonally adjusted data. It’s complex: “One might think that the BEA collects the data on the components [of GDP], aggregates them and then seasonally adjusts the aggregate. For various reasons, however, the BEA instead chooses to deseasonalize the components of GDP,” they write. “Moreover, the BEA does not deseasonalize all the components of GDP. Several economists have independently identified the same residual seasonal patterns in nonresidential structures, government consumption and exports components of GDP, supporting the idea that small seasonalities unaccounted for in the components could be producing noticeable seasonal patterns in the aggregate.”
Another issue is the challenge faced by experts who are trying to make timely assessments of the U.S. economy. A lot of people track GDP data releases: forecasters, business leaders, fiscal policymakers at state, local and federal levels … and, of course, experts at the Federal Reserve.
Owyang and Shell say that “policymakers seeking to make the most timely adjustments to interest rates” could examine other indicators of aggregate economic activity that aren’t subject to the same magnitude of residual seasonality. In particular, they note gross domestic income, or GDI.
Recall that GDP measures activity in terms of what our economy produces, by tracking what is spent on final goods and services. GDI measures activity in terms of what our economy earns, such as wages, profits and interest payments. The authors said that the BEA has tested both indicators for residual seasonality and found that while GDP does exhibit residual seasonality, GDI does not.