Seasonally adjusted

What Is Seasonally Adjusted?

"Seasonally adjusted" refers to economic data that has undergone a statistical process to remove predictable, recurring fluctuations that occur at roughly the same time and with similar magnitude each year. This method, a core component of economic data analysis, aims to reveal the underlying trends and cyclical patterns in a time series by filtering out the "noise" of seasonal events. These seasonal events can include weather changes affecting industries like construction, major holidays influencing retail sales, or academic schedules impacting employment figures. By presenting data as "seasonally adjusted," analysts and policymakers gain a clearer picture of fundamental economic activity, allowing for more accurate comparisons of data from month to month or quarter to quarter. The U.S. Bureau of Economic Analysis (BEA) states that seasonal adjustments ensure movements in economic series better reflect cyclical patterns.¹⁰

History and Origin

The practice of seasonal adjustment gained prominence in the 20th century as governments and statistical agencies began collecting and publishing more detailed economic data. The need arose because raw data often presented misleading spikes or troughs due to predictable seasonal patterns, obscuring the true underlying state of the economy. For instance, before seasonal adjustment, economists analyzing employment data would see a sharp drop in January as temporary holiday hiring ceased, making it difficult to discern if actual job losses were occurring beyond the usual seasonal decline. The U.S. Census Bureau developed the X-11 and later the X-12 and X-13 ARIMA-SEATS methods, which became widely adopted statistical procedures for performing seasonal adjustment. These techniques help government agencies, such as the Bureau of Labor Statistics (BLS) and the BEA, to smooth out these variations. The Bureau of Labor Statistics, for example, uses seasonal adjustment to portray employment and unemployment levels more accurately by removing influences of seasonal events like holidays, weather, and school schedules.⁹,

Key Takeaways

• Seasonally adjusted data removes predictable, recurring seasonal fluctuations from economic time series.
• The primary goal is to highlight underlying economic trends and cyclical patterns, making data comparable across different periods.
• Major government statistical agencies, such as the U.S. Bureau of Labor Statistics and the U.S. Bureau of Economic Analysis, routinely publish data in seasonally adjusted form.
• Examples of seasonal influences include holiday shopping, agricultural cycles, school schedules, and weather patterns.
• While crucial for long-term trend analysis, "seasonally adjusted" data may obscure specific short-term seasonal insights that raw data could provide.

Interpreting Seasonally Adjusted Data

Interpreting "seasonally adjusted" data involves focusing on the core, non-seasonal movements within an economic indicator. When a figure, such as Gross Domestic Product (GDP)) growth or unemployment rates, is reported as seasonally adjusted, it implies that statistical methods have accounted for typical increases or decreases associated with specific times of the year. For example, if seasonally adjusted unemployment rises in December, it indicates that the increase is more significant than the normal, expected seasonal rise in unemployment that often occurs after the holiday shopping season. Conversely, a seasonally adjusted decline in employment in July, a month when factories typically retool, would suggest a deeper reduction in activity than the usual summer slowdown. This adjustment allows for a more accurate assessment of the overall health and direction of the business cycle by separating fundamental changes from routine seasonal variations.

Hypothetical Example

Consider a hypothetical country, "Econoland," which produces a significant portion of its annual output in agricultural goods. Raw monthly GDP figures for Econoland would show a massive spike in GDP during the harvest months of late summer and early fall, followed by significant drops in winter. If Econoland's Ministry of Economy reported unadjusted GDP, it would be difficult to tell if the economy was genuinely growing year-over-year, or if a particular month's high figure was simply due to the harvest season.

To illustrate, let's say Econoland's raw GDP in September was $500 billion. The previous year's raw September GDP was $480 billion. On the surface, this looks like growth. However, after seasonal adjustment, accounting for the typical harvest boost, the "seasonally adjusted" GDP for September might be reported as $450 billion. If the seasonally adjusted GDP for the previous August was $445 billion, this would indicate a steady, underlying growth trend of $5 billion, rather than the dramatic $20 billion spike seen in the raw data. This allows economists to track the actual underlying performance of Econoland's economy, separate from the predictable agricultural cycle. Companies would also use this type of analysis for internal forecasting of sales or production.

Practical Applications

Seasonally adjusted data is ubiquitous in finance and economics, serving as the bedrock for informed decisions by governments, businesses, and investors. Central banks, like the Federal Reserve, heavily rely on seasonally adjusted figures when formulating monetary policy. For instance, when the Federal Open Market Committee (FOMC) assesses the labor market, they look at seasonally adjusted employment data to understand the underlying hiring trends rather than temporary fluctuations caused by holiday hiring or summer layoffs. This helps them determine whether current economic conditions warrant changes to interest rates or other policy tools.⁸

Similarly, analysts in investment firms use seasonally adjusted data for a variety of purposes, from evaluating company performance to predicting market movements. For example, a retailer's sales figures will naturally surge during the holiday season. To assess the true health of the business and its growth trajectory, analysts will examine seasonally adjusted sales to see if the company is growing beyond the expected seasonal boost. The Bureau of Economic Analysis (BEA) uses seasonally adjusted data for its estimates of GDP, allowing policymakers to see true patterns in economic activity.⁷ This practice helps to avoid misinterpretations that could lead to misguided investment strategies or regulatory decisions, ensuring that policies are based on fundamental economic shifts rather than ephemeral seasonal patterns.

Limitations and Criticisms

While essential for revealing underlying trends, seasonal adjustment is not without limitations or criticisms. One significant challenge arises during periods of extreme economic shocks or unforeseen events. The statistical models used for seasonal adjustment rely on historical patterns, and when these patterns are disrupted by events like a global pandemic or a major natural disaster, the adjustments can become less reliable. For example, during the COVID-19 pandemic, the unprecedented shutdown and reopening of economies caused significant distortions. Some critics argued that traditional seasonal adjustment methods struggled to accurately filter out these non-seasonal, yet temporary, shifts, potentially leading to misinterpretations of the true economic picture.⁶,⁵

Another area of concern is "residual seasonality," which occurs when a seasonally adjusted series still exhibits some predictable seasonal patterns, indicating that the adjustment process was incomplete. This can happen if seasonal patterns evolve over time in ways that the statistical models do not fully capture, or if the underlying data quality presents challenges. Research from the Federal Reserve Bank of San Francisco has discussed residual seasonality in official GDP and inflation data, highlighting that despite attempts to adjust, some calendar-based fluctuations can remain.⁴ Such issues can complicate accurate statistical analysis and could potentially, though generally not significantly, influence policy discussions if not properly understood. Agencies continuously review and update their seasonal adjustment procedures to account for changing patterns, but inherent complexities remain.³

Seasonally Adjusted vs. Not Seasonally Adjusted

The key difference between "seasonally adjusted" and "not seasonally adjusted" data lies in their treatment of predictable, recurring annual fluctuations. Not seasonally adjusted data, often referred to as raw data, reflects all movements, including those caused by seasonal factors such as holidays, weather, or school calendars. For instance, unadjusted employment figures typically show a sharp decline in January due to the end of temporary holiday hiring.

In contrast, seasonally adjusted data attempts to isolate and remove these predictable seasonal influences, presenting a clearer picture of the underlying trend. If a January employment report shows a decrease in seasonally adjusted employment, it signifies that the actual decline was larger than the typical seasonal downturn expected for that month. While not seasonally adjusted data can be useful for understanding the magnitude of seasonal shifts themselves (e.g., how many temporary jobs are added for the holidays), seasonally adjusted data is crucial for comparing economic performance across different periods, allowing economists and policymakers to gauge the fundamental direction of the economy without being misled by regular seasonal "noise."²,¹

FAQs

Why is economic data seasonally adjusted?

Economic data is seasonally adjusted to remove the influence of predictable, recurring seasonal patterns. This allows analysts to identify the true underlying economic trends and cyclical movements, making it easier to compare data from month to month or quarter to quarter without distortion from events like holidays, weather, or school breaks.

What kinds of economic data are typically seasonally adjusted?

Many common economic indicators are routinely seasonally adjusted, including employment figures, retail sales, inflation rates, industrial production, and Gross Domestic Product (GDP). Data series that exhibit strong and predictable seasonal patterns are prime candidates for this process.

How does seasonal adjustment help in understanding the economy?

Seasonal adjustment helps in understanding the economy by smoothing out temporary, predictable fluctuations in data. This data smoothing makes it easier to spot genuine shifts in economic activity, such as growth or contraction, that are not merely a result of the time of year. It allows for more accurate assessments of the overall economic climate.

Can seasonal adjustment hide important information?

While primarily beneficial, seasonal adjustment can sometimes obscure specific details, especially if an unusual event coincides with a normal seasonal pattern. For example, if a massive, non-seasonal event causes a large dip in activity during a month that typically sees a seasonal decline, the adjustment might make the dip appear less severe than it was in raw terms. For certain analyses, such as assessing peak holiday hiring, not seasonally adjusted data might be more appropriate.

Who performs seasonal adjustments on economic data?

Government statistical agencies are the primary entities that perform seasonal adjustments on the economic data they collect and release. In the United States, examples include the U.S. Bureau of Labor Statistics (BLS), the U.S. Bureau of Economic Analysis (BEA), and the U.S. Census Bureau. They use sophisticated statistical software and methodologies to apply these adjustments.