The labour shortage has suddenly become one of the main problems of the Russian economy. Government contracts have sharply increased the demand for labour in the defence sector and manufacturing industry. The shortage of workers has provoked frantic demand and wage growth, which is spreading to adjacent sectors and contributing to the overheating of the economy and inflation.
The specificity of the Russian labour market model, which emerged in the post-Soviet period, exacerbates this situation. It is based on three characteristics: low unemployment benefits, low minimum wages and high redundancy costs. This model ensures low formal unemployment and allows the market to adapt well to crises. This is achieved through a large share of informal, low-paid and inefficient employment, which absorbs potential unemployment.
This model is characteristic of many emerging markets and is the opposite of the model of developed economies, where a high minimum wage and benefit threshold leads to higher unemployment but allows low-performing jobs to be weeded out of the market. This model facilitates the structural adjustment of the economy, whereas the Russian model tends to hinder such restructuring. At first glance, discussing unemployment at a time of labour shortage seems nonsensical. In reality, however, unemployment and the accompanying institutional configuration of the market ensure more efficient relocation of the labour force in the context of certain structural or conjunctural shifts. However, the transition to such a model has significant social and political costs and is therefore unlikely.
This publication was prepared by Re:Russia in cooperation with the Academic Bridges project based on the seminar 'The Economic Model of Modern Russia'.
In the 2000s and 2010s, the main problem for the Russian labour market was the inability of the economy to create high productivity jobs: employment was high, but a significant portion of it was unproductive. This low-productive employment, which functioned as a 'sponge' occupying a large share of the overall structure of the workforce, maintained the state of the labour market without unemployment or with low unemployment.
Today, the main problem is considered to be a shortage of labour. Everyone is talking about it: employers, officials, experts. What does this mean? There are difficulties in hiring workers because there is neither unemployment nor reserves outside the workforce. Moreover, there is nothing peculiarly Russian about today's 'deficit' of workers; all developed countries are facing this problem. The post-Covid recovery of economies has created a demand for additional labour for which there is no corresponding supply. This is reflected in the increase in the number of vacancies that are difficult to fill, leading to additional inflationary pressure. For example, in American cities, it is common to see signs in store windows, hair salons, various offices, restaurants, and bars saying 'We're hiring', 'Workers needed’.
Of course, the specific details of the emergence and manifestation of this shortage are different everywhere. I want to take a somewhat detached and external look at the situation on the Russian labour market. We have very few reliable figures to rely on; those figures that do exist are often difficult to interpret. To understand what lies within them requires additional research, which is almost impossible right now.
What is the backstory to what we are witnessing today? Why is there no labour pool? Let us start with recent history. In the 1990s, a specific model was formed in the country, which we called the 'Russian labour market model'. Its prerequisites were established in the late 1980s during attempts to revive 'self-financing' for state enterprises. They were given more freedom to provide material incentives for workers, but not to make surplus staff redundant. This model solidified in the 2000s and continued into the 2010s. Its essence is very simple: wages are extremely flexible, and employment is rigid, and this is the exact opposite of what exists in developed market economies, where wages are rigid and employment is somewhat flexible. However, this reverse flexibility through the adjustment of labour costs ensured the labour market adapted to all shocks, keeping unemployment at a low level. And if unemployment did spike, it would quickly decrease again. All crises in the last 30 years have been handled by the labour market in this way.
This model is based on a specific institutional configuration. First, there is a low wage threshold through low unemployment benefits and a low minimum wage, which provides downward flexibility. Wages can go down because the threshold is low. Second, the variable part of earnings can go up. This includes not only official bonuses, premiums, and allowances but also under-the-table wages. Thus, wage distribution can stretch like elastic in both directions. Third, the rigidity of employment in companies was ensured by the high financial and administrative costs of mass layoffs. Adaptation through individual 'kicking out' takes time and is not of a mass nature.
Thus, this model minimised unemployment by stretching low-paid employment. This includes all kinds of casual employment, temporary employment, part-time employment, semi- and informal employment. Overall, these are low-paid and socially unprotected jobs. Even if they are legal and registered jobs, they exist outside of labour legislation, which significantly contributes to costs. The behaviour of individuals within this model is determined by the fact that low unemployment benefits do not allow them to search for work for long, thereby shortening the job-search period (i.e., unemployment). If a person loses their job, they need to do something for income — give someone a ride, wallpaper someone’s house, take care of someone, help someone out. This market absorbs the employment for which no quality jobs are found and partially replaces unemployment.
The structure of the economy is essentially two-layered. One layer consists of large and medium-sized enterprises, as well as public sector employees. For them, there is a Labor Code and some rules defined within it. Rosstat provides data in its monthly reports specifically for this segment, which accounts for approximately 32 out of over 70 million workers. The other segment includes everyone else — from small businesses to informal self-employed individuals, for whom the Labor Code is a complete abstraction. Both segments include numerous mass professions with easy entry: drivers, salespeople, couriers, security guards — jobs that are relatively easy to obtain, require no special skills, and have become very numerous. These jobs have high turnover and, consequently, frequent new hires.
This model is not unique — in many low- and middle-income countries, we see something similar. The International Labour Organisation (ILO) is well aware of this phenomenon, and its documents note that standard indicators of unemployment and employment are poorly applicable to countries where social protection is weak or absent. Accordingly, statistics should somehow account for this, but they do not fully do so. With this model, achieving unemployment as a political goal is almost automatic. We get full employment and minimal unemployment. Here one might say 'hooray!'. But the situation has changed, and 'hooray' is now followed by a question mark.
In 2020, COVID-19 struck, and what we are witnessing now is difficult to understand and interpret if we forget about the effects of COVID. When the pandemic began, unemployment was already low. During the pandemic, there were no mass layoffs, as was the case almost nowhere in the world, except in the United States. But there was an outflow from the market: for example, people retired or left the workforce altogether.The hiring of new workers was frozen, while latent demand for labour remained but was not satisfied. COVID passed, recovery growth began, and it turned out that there was no free labour supply on the market because unemployment was low and there were also no potential reserves outside the workforce (the employment level was very high). Vacancies and wages started to rise, which became noticeable as early as 2021. However, wage growth across sectors and enterprises with full employment was uneven: those who are able to pay more lure workers away from less financially solvent neighbours. So, when you hire an unemployed person, you thereby eliminate a vacancy and reduce unemployment. And when there is no unemployment and no available labour reserves, by luring away someone and closing one vacancy, you immediately create another. Given that turnover of labour has always been very high in many companies and industries (trade, construction, transportation), the process of 'poaching' leads to further inflation of this turnover and the creation of new vacancies. This, in turn, leads to companies, which witness this frenzy in the market and the threat of worker poaching, creating not one but several vacancies after someone leaves. In essence, wage growth with full employment leads to the creation of new vacancies. The threat of any deficit (including labour shortage) leads to frantic demand — this is a general pattern. Vacancies remain open, and their number increases — it is an endless race until wage growth hits the ceiling of budget constraints.
The situation over the past two years has further exacerbated the problem. As we know, labour demand began to be inflated by government funds in certain sectors requiring a lot of medium-skilled labour. A quick structural manoeuvre in the economy was required. But this structural manoeuvre has been limited by labour supply, and mobilisation, emigration, and reduced immigration have further exacerbated the situation. As a result, we see widespread complaints about a shortage of workers.
The new demand and corresponding poaching of workers predominantly come from the segment of large and medium-sized enterprises in manufacturing (which are somewhat significant). This pool of enterprises is relatively small in terms of the number of employees. However, an industrial enterprise is unlikely to hire workers en masse who previously worked as loaders in a store. It still needs people with certain skills. It is possible, of course, to fill some gaps with temporary work, labour mobilisation, business trips, etc., and companies are very adept at inventing various adaptation methods. But structural shifts imply mass relocation, which is impossible without a free pool of labour, i.e., without noticeable frictional and structural unemployment and without accompanying mass retraining. However, transitioning to such a model of labour mobility — through unemployment and retraining — is politically, institutionally, and technologically quite complex.
From Rosstat's monthly data, we see that over the past two years (from November 2021 to November 2023), the average number of employees in all large and medium-sized enterprises has hardly increased, showing growth of only 0.5% over this period. In other words, employment, despite everything, remains roughly stable. If we consider manufacturing industries as a whole (notably, in the segment of large and medium-sized enterprises, manufacturing industries make up a relatively small sector—with slightly more than 5 million people employed), the growth is 4.4%. If we isolate the sub-sectors that are likely related to the defence industry, this growth is 9%. This data indicates a structural shift, but the source of this additional employment and the actual scale are unclear.
Additional uncertainty arises from the fact that Rosstat uses a rather tricky concept of 'average headcount'. This is the number of people adjusted for the number of hours worked. If two people work one and a half shifts, then the reporting will show three people, not two. We do not know to what extent the growth in employment is associated with an increase in the number of hours worked by the same worker. But, most likely, such an effect is present because in the conditions of labour shortage, the first way to increase labour reserves is to increase the duration of working hours. However, there are limitations, both physical and financial: people cannot work indefinitely, and overtime hours cost more. As a result, companies often use overtime workers without showing this in their reports. But, if we look at the dynamics of hours worked, we will see that in the sectors likely associated with the defence industry, the number of working hours has increased. Although it is difficult to say how many of them are underreported.
To address the labour shortage problem in favour of industries that are currently deemed most crucial by the authorities and in need of a workforce, it is necessary to have available labour reserves. These reserves can only be obtained through some level of unemployment because there are no other reserves. But creating unemployment within the existing labour market model is impossible due to institutional and political constraints. This appears to be an almost insurmountable problem.
The demographic situation in Russia continues to deteriorate, which means that the constraints on the labour market are increasing. These restrictions are not only reflected in the declining working age population but also in what R. Kapeliushnikov termed the 'youth dilution' of employment, i.e., the rapid decrease in the number of young workers in the workforce structure. For example, information technology, which is crucial, is dominated by young people. The main structural manoeuvre in the economy today is associated with a return to labour-intensive technologies, which typically require a large number of physically healthy individuals with medium qualifications, who, as we see, are lacking.
If the money for salaries runs out, poaching workers will become more difficult, and the money will run out sooner or later. Even despite the budgetary boost received by companies, further wage increases are becoming increasingly difficult. And this resource for shifting labour will also be depleted. The only remaining option in this situation is the forcible compression of the second segment of the labour market, which is outside of large and medium-sized enterprises (I am not talking about such a reserve as the increase in labour productivity here; it is much easier to discuss this than to achieve it). And now there are discussions about there being too many self-employed individuals, what they are doing, and how much they are paying in taxes. From here, ideas might emerge on how to clamp down on these self-employed individuals to make them work in factories and plants.
In general, it can be said that the institutional model that ensured full employment and allowed the government to boast about its successes in combating unemployment, which was lower than anywhere else in the world, has become a specific and serious constraint today. It poorly coexists with the absence of a pool of available labour. The structural manoeuvre observed in manufacturing, and the outflow of labour, only exacerbate the challenges it faces. How it will cope with them, I dare not predict.
It may seem that in a situation of stark labour shortage, even the theoretical assumption of increasing unemployment seems completely absurd. However, if one of the key institutional features of the Russian model is the low wage threshold, then, for example, a significant (two or threefold) increase in unemployment benefits, combined with easier access to them, may significantly push up the wage equilibrium (but there are other levers for this threshold). Incentives to stick to low-paid jobs would weaken. The result would be an increase in the average and minimum wage, accompanied by the release of those who cannot reach the new wage minimum in terms of their productivity.
This was the path taken by the countries of Central and Eastern Europe at the beginning of the 1990s, when faced with the task of restructuring their economies. A relatively high level of social protection (unemployment benefits) squeezed out unproductive workers from their jobs and became one of the factors of high unemployment in the early transition period. This accelerated the structural adjustment of the economies of these countries, whereas the Russian model contributed to the adaptability of the labour market to various crises but tended to hinder restructuring. Thus, the boundary between deficit and unemployment is not a concrete wall but depends, all other things being equal, on the adopted institutional configuration of the labour market, which in turn has much broader implications for the economy than is commonly believed.