Employment at a decades-high yet no room for dynamic growth

This article is a part of Poland Unpacked. Weekly intelligence for decision-makers

The medium-term view

From a medium-term perspective, however, employment has remained broadly stable. In the fourth quarter of 2022, the number of people in work stood at 17.32 million – only marginally lower than today, by around 30,000. The increase recorded in 2025 has more than offset last year’s decline, leaving a small net gain.

This is, of course, largely a demographic story. With the size of the population aged 15–89 constraining labor supply, and unemployment low – at 3.2% in Q4 2025 – there is limited scope for further growth, beyond gains driven by higher labor-force participation. The latter category includes both those in employment and the unemployed who are actively seeking work. The remainder are classified as economically inactive.

Changes to the labor force survey

Labor Force Survey (BAEL) results change over time. One reason is the rebenchmarking that takes place every ten years following each population census. Data are revised retroactively for several years, though not for the entire historical series. Over longer horizons, this leads to breaks in continuity, complicating comparisons over time.

A second reason lies in numerous changes to the survey itself. For example, since 2021 the survey has covered the population aged 15–89 (previously 15+, with no upper age limit). Definitions have also been adjusted. Individuals who are self-employed in individual farming and whose output is intended mainly for own consumption – while having no other job – have been excluded from the category of the employed.

There have been further changes, both in 2021 and in earlier years. Two economists, A. Dormitz and O. Zajkowska, recently published in Materiały i Studia of National Bank of Poland (NBP) a proposal to adjust historical data for 1995–2020 so as to ensure definitional consistency with the current version of the survey. Without delving into the technical details of this valuable work, it is sufficient to note that the analysis here will rely on the revised historical data.

Record labor-force participation and employment

Labor-force participation currently stands at 59% - the highest level since 1995, based on the harmonized data series.

The relatively high participation rates seen in the mid-1990s (for example, 56.7% in Q3 1995) masked a comparatively low number of people in work. The reason was very high unemployment, which at the time reached 2.2 million. By contrast, the latest data put the number of unemployed at 570,000.

This is clearly illustrated in the chart below, which shows the absolute number of people in employment between 1995 and 2025. Thanks to much lower unemployment and slightly higher labor-force participation, the number of people in work today is around 3 million higher than in the mid-1990s – an increase of roughly 20%.

Have the best years already passed?

Even so, the period of the strongest increases in employment appears to be behind us. This is shown in the chart below, which presents changes in the number of people in work over rolling five-year periods starting in the fourth quarter of 1995.

In the most recent five-year period, ending this quarter, employment rose by 380,000 employees and employers. That is more than four times less than during the peak growth years of 2005–10, when the number of people in work increased by almost 1.7 million. An unfavorable shift in the size of the working-age population, combined with already low unemployment and a slowdown in immigration, means that future gains will have to come primarily from higher labor-force participation. Yet without measures such as reform of the retirement age, the impact of this factor — while positive — will remain limited.

A final caveat

It is worth noting that during the pandemic the Labor Force Survey (BAEL) also underwent other significant methodological changes. These included a shift in the mode of interviewing – from face-to-face to telephone – as well as revisions to the questionnaire itself. According to another study by the already cited O. Zajkowska, these changes increased the estimated labor-force participation rate by 0.6 percentage points. The adjustment used in my analysis does not take this effect into account.