Reserves of Russia’s labor market

GENERAL CONCLUSIONS

• The Russian labor market is in a state of stress, although in the global context this situation is far from unique. In Russia and many other countries, this situation is related to demographic trends, the legacy of the pandemic, as well as a sharp change to the structure of demand for labor. The degree of overheating of the market can be estimated at 1% of the workforce, or 700,000–750,000 employees.

• The Russian labor market has reserves, but they are not contained in the unemployment rate. Among the economically inactive part of the population there is a large number of people who want to work, but cannot due to various reasons. In addition, the desire to work itself may depend on labor market institutions and government policy. ACRA assesses internal potential for additional supply of labor resources at 7.5 mln people if in each of the age groups, economic activity grows to levels that are characteristic of Russia’s peer group of countries. The use of this resource will probably require the state to employ a policy of stimulus in the labor market and change institutions with a view to increasing the flexibility and diversity of labor relations.

• Trade is experiencing relatively high and consistent stress in terms of hiring compared to other industries. It seems this applies to different segments — from car sales to wholesale and retail trade in other goods. The situation in other sectors is changing rapidly, periods of stress alternate with calmer periods, and stress in general manifests itself unevenly not only by industry, but also by professions and companies.
  

LABOR MARKET UNDER PRESSURE

According to the established narrative, the Russian labor market has been overheated for at least the last year and a half. This is indirectly evidenced by record low unemployment and growth of real wages (by 7.5 pps over 2022–2023) amid declining average labor productivity (employment grew by 2.6–3%, while real GDP increased by 2.3% over the aforementioned period). It is also customary to say that the ‘employer’s market’ has become an ‘employee’s market’. Various surveys of employers confirm this conclusion¹.

The employment of the unemployed is historically the most important, but actually an exhausted reserve. According to Rosstat’s latest data, only 2.9% of the workforce is not employed in the Russian labor market, and the unemployment rate in 2023 averaged 3.2%. Are these small figures or large ones? They are undoubtedly small —there are currently five to ten countries, mainly industrialized and highly dependent on immigration, that have similar conditions. Another way to assess the small size of observed unemployment is to compare the indicators with the model ‘equilibrium’ levels specifically for Russia. There are two common approaches to determining this value — the ‘natural’ level and NAIRU (non-accelerating inflation rate of unemployment). Without going into detail on methodology, we note that the multitude of assessments for Russia, as well as ACRA’s assessments, result in both equilibrium levels exceeding 4%². Therefore, with the first approximation, we receive a quantitative assessment of the degree of overheating of the market — at least 1% of the workforce (this is a sort of anti-reserve of the labor market).

¹ Rosstat’s data on the business climate in Russia shows a significant increase in the number of positive responses from respondents to the question of whether the shortage of labor restricts growth of production in the manufacturing industry (an increase from 23% to 29% over the past two years) and in the service sector (increase from 26% to 34%). According to a GIEP market survey, in January 2024, 47% of managers of industrial enterprises complained about a lack of qualified personnel, which is a historical maximum for this survey that has been conducted since 1996.

² For an overview of work and approaches, for example, see: D. Orlov, Phillips Curve: Inflation and NAIRU in Russian Regions / D. Orlov, Y. Postnikov // Zhurnal Novoi Ekonomichesky Assotsiatsiya — 2022 — No. 3 (55). — pages 61–80. DOI: 10.31737/2221-2264-2022-55-3-4.

Population of a country = economically active population (workforce) + economically inactive population.

Economically active population (workforce) = employed + unemployed.

Figure 1. Distribution of countries* by unemployment rate, % of workforce

* 60 largest economies of Europe, Asia, North America, and South America.
Sources: International Labor Organization, national statistics agencies, ACRA

The natural decline of the population is the excess of the number of deaths over the number of births.

The reasons why the labor market in Russia has been in a tense state for several years are not unique in many ways.

1.    Demographics. If the natural decline of about 500,000–800,000 people per year is not compensated by immigration, then it sets the trend for a noticeable decrease in the population³. Even taking immigration into account, over the next five years, the number of people aged 20 to 39 (statistically the most productive cohort, as well as the main source of new employees entering the market) will decline by 4–5 mln⁴. The number of people aged 15 to 65 will fall by 2.5 mln over the next five years. Consequently, the potential supply in the labor market is also declining.

2.    Legacy of the pandemic. In 2020, amid lockdowns and other crisis phenomena, the rates of both hiring and dismissal declined noticeably. As a result, delayed turnover of employees formed, which turned out to be not as easy to restore and increase as quickly as the recovery in demand for the companies’ products suggested.

3.    Changes to the structure of demand for labor. In the world as a whole, the shortening of production chains, which is accompanied by corporate investment in relocating production facilities from other jurisdictions, leads to a change in the list of the most needed professions and increases demand for labor because it reduces the return on scale in industries and companies undergoing the reorganization process. In addition to these trends, the demand structure in Russia has shifted toward the defense industry and military service. The structure of the labor supply in terms of professions and competencies is changing much more inertially.

In addition to these reasons, sometimes there is a long-term change in employee preferences regarding working conditions, excessive accumulation of labor resources by companies for precautionary purposes, etc.⁵

³ The data from the demographic forecast of Rosstat is presented, which relates to all three of its variants: Главная страница/Статистика/Официальная статистика/Население/Демография/Демографический прогноз.
⁴ Medium version of Rosstat’s demographic forecast
⁵ For example, see: B. Van Doornik, D. Igan, E. Kharroubi. Labour markets: what explains the resilience? BIS Quarterly Review, December 2023.

LABOR MARKET RESERVES: WHERE WILL NEW SUPPLY COME FROM?

There are many indicators that can be used to diagnose the state of the labor market. They are overshadowed by two of the most popular indicators (unemployment rate and wage dynamics), but contain no less useful information and indicate theoretically existing hidden reserves. The creation of incentives to use these reserves could become a target of economic policy.

Internal reserves for growth of employment

1.     Potential workforce and others who wish to work. An adult is classified as unemployed if they meet the following formal criteria: (1) are not currently employed, (2) have searched for work in the past month, (3) are ready to start working within a week⁶. Failure to comply with at least one of the criteria listed under numbers two and three leads to a person being excluded from both the unemployed and the workforce as a whole. In Russia, the number of these people is now higher than the number of unemployed in 2022 and amounts to 4% of the workforce. At the same time, practice shows that it is possible to move from these intermediate states to a state of employment⁷.

If we consider people who do not meet the second or third criteria to be a reserve of the labor market, for which, using public policy, it is possible to simplify the search for work and stimulate willingness to work, then it makes sense to use the indicator of the extended unemployment rate (Fig. 2, green line). This is the ratio of people who are not employed but want to work (regardless of whether they are looking for a job and are ready to enter it soon) to the size of the workforce. This ratio is currently 6.9%, which is more than twice the unemployment rate calculated according to sample surveys published by Rosstat, and an order of magnitude higher than the registered unemployment rate (2.9% and 0.6%, respectively).

⁶ See Rosstat’s website for a detailed description of the criteria: Главная страница/Статистика/Официальная статистика/Рынок труда, занятость и заработная плата/Трудовые ресурсы, занятость и безработица/Занятость и безработица/Методологические пояснения.
⁷ V. Gimpelson and A. Sharunina in the article “Flows in the Russian Labor Market: 2000–2012” indicate that for the entire economically inactive population (not only for those who wish to work) the possibility of being employed in the next year is 16% vs. 51% for the unemployed. It is logical to assume that among those who wish to work, which are mentioned in this ACRA research, this probability is significantly higher than 16%. 

Not all components of the extended unemployment rate are included in the labor force, so this indicator cannot be interpreted as the share of the labor force potentially becoming employed; this is precisely the ratio. In an alternative calculation, an expanded definition of labor can be used in the denominator of each of the indicators in Fig. 2, but we want to keep the standard indicators unchanged, while ensuring comparability with the expanded indicator (the same denominator).

Figure 2. Unemployment indicators in Russia, % of workforce

* SES — State Employment Service
Sources: Rosstat, ACRA

2.    Part-time and remote jobs and other flexible forms of employment. Not all people included by ACRA in the calculation of the extended unemployment indicator (people who do not work but want to work) can become employed in the current conditions. There are various reasons for this situation: health status, marital status, age, educational level, competencies that do not match the current market requirements, and working conditions that do not meet expectations. The same reasons may explain the lack of desire to work among a different part of the economically inactive population. Some problems could be solved if flexible forms of employment (part-time, remote, platform-based, etc.) become more widespread in the market. Flexibility is of course perceived differently by employees and an employers, but ultimately is about having more forms of employment and increasing the probability of using the labor potential more effectively.

Part-time employment is one of the flexible forms of employment. Currently, about 1% of employed Russians work part-time on a contractual basis. Using internal reserves of the labor market may increase this proportion noticeably. On the contrary, non-contractual part-time work (forced downtime, ‘voluntary’ leave, etc.) is a phenomenon that rather indicates crisis adaptation of the economy and, along with a drop in real wages, is the Russian equivalent of growing unemployment in other countries. The use of labor market reserves helps reduce this share.

Remote employment, a phenomenon that affected more than 20% of employees at the beginning of the pandemic, now affects just several percent of the employed. However, the experience gained during the pandemic has increased the willingness of companies and employees to maintain these labor relations. The use of labor market reserves is likely to increase the share of remote jobs in the long term.

Figure 3. Part-time employment in Russia, share of employed

3.     Pension reform and employment for the elderly. The transition period of the pension reform, which began in 2019, will last until 2028. The reform’s essence is to raise the retirement age for both men and women by five years. In 2024–2028, the number of working age people will increase by 3–4 mln compared to the ‘no reform’ scenario, which is comparable to 5% of the workforce (since the beginning of the reform, the increase will be about 8 mln people, or slightly less than 10% of the workforce). However, it is incorrect to assume that this indicator represents the entire labor market reserve and the number of employees may increase by this amount due to later retirements, since about half of the Russian retirees continued to work for at least five years after their retirement age even before the reform was implemented. Statistics of the first years of the transition period show that the real effect of raising the retirement age is an increase in employment of the age groups directly affected by the reform by only 10–15 pps⁸ (and not by 100 pps, as one might think). This is comparable to a potential increase in the number of employed by a little more than 1 mln people over the entire transition period (about 1.5% of the workforce).

If the Russian labor market begins to use its internal reserves more effectively, statistics will reflect it as an increase in the share of employed out of the total population. To what level could this share grow? We can try to give an upper bound estimate based on global experience. Among countries similar to Russia in terms of the duration of education and mortality rates, higher employment rates are observed in some age groups. If we assume that Russia can achieve similar indicators as a result of public policy, incentive measures, and changes in labor market institutions toward more flexible labor relations, then the internal reserves of the labor market can be estimated at 7.5 mln people, including about 1.5 million young and almost 5 million elderly people⁹ (Fig. 4). At the same time, the pension reform’s contribution to the potential involvement of the latter group in the labor market seems to be small.

⁸ R. Kapeliushnikov. Russian labor market: statistical portrait on the backdrop of crises / R. Kapeliushnikov // Voprosy ekonomiki — 2023. — No. 8. — pages 5–37.
⁹ A similar method of estimation (with other target countries) is described in the report “Russian labor market: trends, institutions, structural changes” issued by the Labor Research Center and the Labor Market Research Laboratory of HSE University in 2017 (edited by V. Gimpelson, R. Kapeliushnikov and S. Roshchin). Depending on initial assumptions, this report estimates the potential employment growth through involvement of older ages at 670,000 to 6 mln people.

Fig. 4 shows the percentage of the employed Russian population in each age group. Green rectangles are a range in which similar proportions are found in all countries. The red line is the median for them. Gray rectangles show a similar spread in countries comparable to Russia in terms of mortality rate and education period. Arrows show a hypothetical increase in employment in Russia to the maximum levels of comparable countries. 

Figure 4. Employment potential by age

The external reserves of the labor market are based on immigration. The moderate version of Rosstat’s demographic forecast assumes that over a 20-year horizon, the net migration inflow will compensate for about a third of the natural population loss and will decrease by about 20% relative to the 2010s (to 190,000–200,000 immigrants per year). The prerequisites for a decrease in migration inflows seem reasonable¹⁰, since in Russia’s common immigrant donor countries, regardless of the different stages of demographic transition, a slowdown in the growth of population aged 20–45 (the most active age for migration) is inevitable. In addition, in many of these countries, living standards and internal labor demand are growing faster than in Russia (from lower levels), which also constrains the use of external reserves for the Russian labor market. There is also noticeable competition between the immigration recipient countries. Therefore, in order to use the external reserve, it is necessary to pursue an active economic policy that facilitates immigration and adaptation and improves social guarantees; otherwise, the current situation can rather be described as a tense one.

The labor market reserves described above are quantitative, however, the degree of tension in the labor market also has a qualitative dimension. The sectoral structure of labor demand, having undergone various transformations, is not necessarily satisfied with the current supply structure.

¹⁰ Under certain assumptions, there is risk of a more significant decline — by 35–45% vs. the 2010s. These results are shown by the migration inflow simulation model that captures the following factors: difference in well-being, population aged 20–45, and the share of population fluent in the Russian language.

unsatisfied demand: a patchwork quilt of vacancies

Labor demand statistics are more difficult to analyze than labor supply statistics. Currently, demand is changing faster. Available sources of information cover different market segments, sometimes overlapping each other, sometimes leaving gaps. One can refer to the three variants of the so-called Beveridge curve: the first one  is a curve based on data from the State Employment Service, the second is a curve based on the data from hh.ru (one of the largest job search services), and the third is a curve based on Rosstat’s data collected during a survey of large and medium-sized enterprises.

Fig. 5 shows the Beveridge curve based on data from hh.ru. The 2019–2023 dynamics show that the labor market cooled down significantly and lost its effectiveness during the pandemic, then it gradually warmed up in H2 2021, after which, apparently, it almost returned to the pre-pandemic level of efficiency in 2022, and  continued to warm up to historical highs by the end of 2023¹¹. In this case, efficiency means, among other things, the extent to which the structure of existing competencies, experience and requests of employees corresponds to the industry profile of demand from companies.

¹¹ For details on interpreting the dynamics of the Beveridge curve, see, for example, O. Blanchard, A. Domash, L. Summers, Bad news for the Fed from the Beveridge Space, PIIE Policy Brief.

Figure 5. Beveridge curve for Russia

The curve based on data from the State Employment Service will reflect an increase in the labor market efficiency by the end of 2023 compared to 2019, while a graph based on Rosstat’s indicators for large and medium-sized enterprises will show its decline. Apparently, this is explained by different coverage of industries by the three data sources, and as a result, it indicates a different state of affairs in these segments of the labor market. The State Employment Service accumulates mostly information on relatively low-paid jobs in the real sector of the economy, whereas the data from the large and medium-sized enterprises are focused rather on basic industries (including manufacture and construction), regardless of the level of wages, while hh.ru shows a lot of service sector vacancies and related resumes. Given these considerations, by the end of 2023, the most tense labor market situation was typical for large industrial enterprises and the construction industry, while low-paid jobs were filled relatively efficiently, whereas the service sector was between these poles.

More fractional data on the dynamics of the ratio of vacancies and CVs reveals a more complex picture. According to hh.ru, the most difficult situation in terms of meeting new demand is in retail trade, the automotive business, and the tourism industry. There are also some problems finding housekeeping aides and blue-collar workers in general¹².

The above conclusions somewhat contrast to Rosstat’s data based on a sample survey of the labor force regarding the number of employed by each industry, as well as to data on the dynamics of average wages (Fig. 6). Among the large industries, the manufacturing and public administration (including social security and military employment) segments demonstrated a combined growth of real wage and employment in the period up to November 2023; the employment rates declined in the construction, electricity supply, agriculture, and mining industries¹³. The greatest tension in the labor market shown in Fig. 6 could look like an increase in real wages against the background of weak employment growth. From 2022 to 2023, as shown in Fig. 6, the situation was tense in the construction, electricity supply, and finance industries, and within the shorter horizon covering only 2023, in the agriculture, trade, finance and public administration (including military employment) segments.

¹² The conclusion is made based on the growth of the hh index by industry (ratio of active resumes to vacancies) for 2023. The stronger the relative fall in the index in an industry, the more tense is the situation in the industry. In ACRA’s opinion, the comparison of initial indexes is less informative since it reflects industry specifics in terms of posting resumes and vacancies on hh.ru.
¹³ A more than 15% fall in the average real wage in the mining industry shown in Fig. 6 is partially a statistic illusion caused by seasonal variation of wages in the sector.

Figure 6. Dynamics of wages and employment by industry from 2022 to 2023

Sources: Rosstat, ACRA

Having compared the contradictory data described above, it is relatively safe to conclude that the trade sector is constantly experiencing  high stress in terms of hiring employees compared to other industries, though this situation seems to be relevant for different segments — from car sales to wholesale and retail trade in other goods. The situation in other sectors is likely to change rapidly; stress periods alternate with calmer periods, and the general tension manifests itself unevenly not only in industries, but professions and companies as well.