It is natural for labour market matching to deteriorate in conjunction with a deep crisis but the failure of matching efficiency to recover after crises is mainly due to changes in the composition of the unemployed. Groups with lower job-finding rates, such as refugees and their family members, saw a relatively large increase in Sweden from the mid-2000s until the pandemic (see Figure 2). Many members of these groups initially lack several of the skills needed to quickly enter the Swedish labour market, such as Swedish language skills, relevant education and work-related networks. The proportion of jobs that do not require a specific education is low in the Swedish labour market and has also declined over time, while the share of people with at most lower secondary education has increased among the unemployed (see Figure 3). In addition, negotiated minimum wages are relatively high in Sweden. For many foreign-born persons, it takes time to become established in the Swedish labour market.^[8] Among refugees and their family members who were received for resettlement by municipalities in Sweden in 2014–2018, 15–20 per cent were employed after one year and about half were employed after 4 years. Consequently, aggregate matching efficiency also declines for quite some time after a large inflow of immigrants with low job-finding rates. When they have been in Sweden longer and have become established in the labour market, matching efficiency should recover.

Figure 2. Immigration by grounds for settlement Thousands of persons

Figure 3. Proportion of elementary occupations and unemployed persons with low levels of education Per cent of all unemployed persons (left axis) and all employed persons (right axis)

The deterioration in matching efficiency over time is mainly due to changes in the composition of the foreign-born population. This can be illustrated by drawing the Beveridge curve separately for Swedish-born and foreign–born persons (see Figure 4).^[9] As it is not possible to separate vacancies for Swedish-born and foreign-born persons, the vacancy rate refers to all vacancies divided by the total labour force. The Beveridge curve for foreign-born persons shifted outwards in 2009 and 2020. One likely explanation for the outward shift in connection with the pandemic is that sectors that employ many foreign-born persons, such as the hotel and restaurant sector, were particularly hard hit by the pandemic restrictions. As a result, unemployment among foreign-born persons increased more than among Swedish-born persons (see Figure 5).

Among Swedish-born persons, there are no similar shifts in matching efficiency but the changes appear to be cyclical movements along the curve. Figure 4 also shows that, in the third quarter of 2024, the Beveridge curve for both Swedish-born and foreign-born persons has reverted to the pre-pandemic relationship as post-pandemic hysteresis effects have faded. Matching efficiency among foreign-born persons has thus improved in recent years. The total Beveridge curve in Figure 1 continues to show slightly worse matching efficiency than in the years before the pandemic because foreign-born persons now make up a slightly larger share of the total number of unemployed people. It is likely that the Beveridge curve is also affected by the weak economic situation and that it may move inwards as the economy improves.

Figure 4. Beveridge curve 2005–2024 for Swedish-born and foreign–born persons

Figure 5. Unemployment among Swedish-born and foreign-born persons, 15–74 years Per cent of the labour force