Permanent employment has been on the rise since 2019, but turnover is skyrocketing and failing to reduce the flow of workers into unemployment.

When new labor market data are published, such as the quarterly updates to the Labor Force Survey (EPA) or data on registered affiliation or unemployment, we often think of the labor market as a mass of workers to whom jobs are added or subtracted depending on the economic cycle. The reality, as usual, is much more complex, and to understand the evolution of the labor market, it is necessary to analyze not so much the number of employed or unemployed, but the flows between employment situations, a set of probabilities that change depending on the type of worker and the general economic situation.

In this regard, the recent work of the Independent Authority for Fiscal Responsibility (AIREF) on changes in the labor market between the years prior to COVID-19 and the 2022/2023 period helps us understand the main transformations in the labor market in Spain based on these flow data. The first relevant data point is the probability of moving from unemployment to employment, a probability estimated based on administrative data from Social Security. This probability is closely related to the duration of the unemployment episode, rising from 14% in the first month to 5.5% after a person has been unemployed for a year. Regarding the trend, the data indicate that, regardless of the duration of the unemployment episode, the probability of finding a job has increased between the 2018/2019 and 2022/2023 periods.

Although the increase in the probability of finding employment is modest, the main change is in the type of employment found. When broken down by contract type, the probability of finding temporary employment after six months of unemployment is almost identical, while the probability of finding permanent employment under the same circumstances has increased from 7% in 2018/2019 to 11% in 2022/2023. In other words, while in 2018/2019, the probability of exiting unemployment into temporary employment was twice as high as that of permanent employment, now the two probabilities are practically the same.

The other side of the coin is reverse transitions, the probability of moving from employment to unemployment or inactivity. In this case, the changes have been much more modest since 2018/2019, and, if anything, a slight increase in this probability is observed across the distribution of workers according to their length of employment. However, although no major changes are observed in aggregate terms, the Airef does document significant changes in the probability of job loss among permanent workers.

In line with other previous studies (see here ), Airef 's results indicate that the probability of job loss has skyrocketed among permanent workers since 2018/2019. Specifically, this probability has increased from 6% during the first month of employment to 11%, from 2.2% in the sixth month to 3.6%, converging to the same levels only after two years of employment. This presents a new, and somewhat unexpected, reality in the labor market.

While the increase in permanent contracts would suggest that the probability of job loss would have decreased in aggregate terms, this transformation among new permanent workers has meant that, despite the greater proportion of permanent workers, the probability of job loss has remained almost unchanged. However, Airef clarifies that, once workers' characteristics (which have changed since 2019) are controlled for, the adjusted probability of job loss has indeed decreased slightly, especially in the first months of employment, although much less than the increase in permanent workers might suggest.

The key to this analysis is to attempt to disaggregate the extent to which these changes are attributable to the 2022 labor reform, which significantly restricted temporary employment in favor of permanent and fixed-term contracts. Using microdata on labor flows from the Labor Force Survey and employing a difference-in-differences model, Airef estimates the impact of the reform on the probability of transitioning between different labor market states as a result of the reform.

In the case of transitions from employment, a negative and statistically significant effect is found on the probability of moving from employment to unemployment, especially among temporary workers, although no effect is found on the probability of changing from a temporary to a permanent contract. In aggregate terms, this means that while overall employment did not grow as a result of the reform, the proportion of permanent workers did, although this did not translate into greater protection against unemployment (measured by the probability of losing their job).

The reduction in the temporary employment rate is one of the major changes in the labor market over the last decade, a change whose effects we have yet to fully understand in many areas. In one of these areas, labor protection, it would be expected that the higher proportion of permanent contracts would have translated into a lower aggregate probability of job loss for workers.

However, increasing analyses indicate that, although temporary employment has decreased on paper, the changes in permanent employment are contributing to the constant presence of "real" temporary employment. In this regard, the latest data from Airef confirm that, despite the substantial increase in the probability of finding permanent employment, the probability of moving from permanent employment to unemployment has increased to such an extent that the probability of losing one's job has remained virtually unchanged since the pre-COVID period.