Seasonally Adjusted Data

Many economic statistics, such as employment, retail sales and production, rise and fall at predictable times of year. Without adjustment, these seasonal swings can hide what is really happening in the economy.

For example, unemployment often rises in January after temporary holiday jobs end. If unemployment rises in January by about the usual seasonal amount, the seasonally adjusted figure would show little change, correctly indicating that the labour market has not really weakened. If unemployment rises much more than usual, the adjusted figure would show an increase.

Statistical agencies use specialised methods, based on patterns observed in past years, to estimate and remove seasonal effects. Adjustments can also account for differences such as the number of working days in a month or the timing of holidays that move from year to year.

Most headline economic figures are reported on a seasonally adjusted basis, so that changes from month to month or quarter to quarter can be compared meaningfully. Seasonal adjustment involves estimates, and figures may be revised as more data becomes available. Unusual events, such as the COVID-19 pandemic, can make seasonal patterns harder to estimate.