In recent years, the increase in income inequalities has become one of the main subjects of social debate. In Spain this debate is especially relevant because it is one of the European countries where inequalities have increased most, above all following the economic crisis of 2008. Economic inequalities are a consequence of the functioning of the markets and economic activity, and of an income distribution that favours some people and groups above others. Governmental redistributive action can alter these inequalities through instruments such as tax policies (so that people who have a higher level of income contribute proportionally more) and through public spending, preferentially benefiting the more disadvantaged groups.

Changes in income distribution are, therefore, a combination of what happens in business activity and the effect of certain public policies. The labour market plays an essential role because this is where primary incomes are established for the majority of the population. 

This study analyses to what extent changes in the labour market have contributed to the increase in inequalities over the last decade. To analyse this question, the best sources of official statistics available have been used. For salaries, this means the Annual Labour Structure Survey. This source is published with a considerable time lag, so we only have access to data up to the year 2016. The other sources used are the National Accounts of Spain and theActive Population Survey (EPA). The basic period under study runs from 2008 to 2016, which corresponds to the period of the economic crisis and destruction of employment (2008-2013) and the first phase of recovery (2014-2016). In cases where information is available for 2017, we have added it to show that the continuity of the recovery has not changed previous tendencies.

The economic crisis has caused a downward pressure on salaries through two complementary routes. Firstly, growth in unemployment always promotes salary moderation, as it implies an increase in the number of people competing for a job. Secondly, austerity policies, applied from 2010 onwards, have led to salary devaluation. This devaluation has been achieved through the labour reform of the years 2010 and 2012, whose measures encouraged the fragmentation of collective bargaining, less protection for employment and an increase in the discretionary power of companies in setting salary levels and job conditions (Pérez Amorós et al. 2016).

In table 1, two elements are combined. One is the share of wage incomes in overall Gross Domestic Income, a measure that makes it possible to show what percentage of income is represented by wage incomes (with the rest being fundamentally business profits and the incomes of self-employed workers). 

Wage distribution between salaries and profits usually remains quite stable, therefore the relevant changes that have taken place, such as the fall of over four points in the 10 years represented in the table, are indicative of important transformations in income distribution. This fall might be due to a transformation in the labour structure, in which salaried employees represent a lower proportion of the occupied population (and, therefore, more people receive non-salary incomes), but the second column of data shows that the weight of wage-earners in the occupied population has remained stable over the course of that same period, and it has even increased slightly in the last time period considered. It should be noticed that this is analysis of primary income distribution and that the unemployed and inactive populations are not taken into consideration with regard to their incomes which, where they exist, originate from redistribution processes.

During the period 2008-2016, the fall in real salaries has occurred in all sectors of activity except for mining and water and sanitation, two sectors that are not very important in terms of global volume of employment. However, this salary reduction has not been even across the board; rather it has varied in each sector depending on its collective agreements and employment contracts.

It is important to point out, furthermore, that this is the average wage for the sector. This means that the differences observed are due as much to wage variations as to changes in the composition of the employment. For example, if a large number of people with high wages retire and new workers are hired with lower salaries, then the average wage falls, while the reverse is true, if new employees earn higher salaries. We should be aware that the Survey on Salary Structure, from which the data have been extracted, does not contain information on the two sectors that traditionally pay the lowest wages and in which informal employment proliferates: agricultural work and domestic service.

The most important salary falls have taken place in activities where low wages are usually predominant (hotel and catering, auxiliary services, other services) and in sectors that were most affected by public cuts (education or healthcare and social services). The only exception to this pattern is the financial sector, where the strong salary adjustment has been associated with significant job destruction especially concentrated among older employees, i.e., those who accumulated relatively high incomes. The increases in nominal wages (absolute amounts paid to employees) reflect the variations in wages that have taken place. The real wage is obtained deducting from this increase the growth in prices (CPI), and it reflects the variation of purchasing power due to changes in salary and prices. Over the course of the period studied, the reduction of purchasing power or increase in inflation was 14% (table 2).

The combined effect of this differential evolution of salaries has led to an increase in wage inequalities. To evaluate these, two measurements will be used, one of which is the Gini coefficient, which offers a synthetic measurement of inequality. Its values are expressed between 0 (non-existence of inequality) and 100 (maximum inequality, an individual would obtain 100% of the salaries and the rest would have none). This coefficient shows an unequivocal tendency towards growth in the last decade, thus indicating that salary inequalities between individuals have grown (graph 1).

The second measurement is the D9/D1 index, which compares salaries of the 10% of the population that earns the highest incomes (D9) with the 10% of those who receive the lowest wages (D1). Deciles (D) are the earning values that, ordered from lowest to highest, divide the number of workers into ten equal parts, so that each of them includes 10% of the workers.This index shows the persistence of an elevated degree of wage inequality, although, in this case, the tendency towards increased inequality is not as clear, perhaps because the adjustment of employment in the financial sector, as we have commented, has fallen upon higher salary levels. It is also possible that the results of the D9/D1 index are affected by the composition of the sample each year (graph 1). In more egalitarian countries, the ratio is close to 2.

Low salaries are associated with worse employment conditions. In Spain, undesired part-time work is one of the circumstances that contribute to social precarity. Following the economic crisis, part-time employment has grown, but this growth is due to the fact that this work is the only available option (graph 2), and not because it is voluntarily sought.

These part-time jobs offer less income due to the fact that fewer hours are worked and they tend to be concentrated among the lowest strata of the salary hierarchy. In fact, in the analysis of salary structure, people with part-time employment represent a majority in the worst-paid 10% of jobs, with their annual incomes below the annual minimum wage. 

Among the part-time employees there is a prevalence of women (graph 3) and growth is more related with the existence of production processes that present short-duration peaks of activity at specific times (for example school catering) than with the search for private reconciliation alternatives. 

With respect to temporary employment, the economic crisis represented a major reduction in temporary contracts, particularly due to the mass destruction of employment in the construction sector where these types of jobs are especially abundant. After the crisis, temporary work grew once more and now stands at levels close to those of the start of the crisis.

Temporary jobs are not only associated with employment instability, but also with lower wages. The reason is the same as that already pointed out in the case of part-time jobs: the concentration of temporary jobs at the lowest levels of the employment hierarchies and the lower number of working hours prevent access to different wage supplements. 

In table 3 we see that, on average, temporary employees earn 25% less than people with a permanent employment contract. Furthermore, resorting to temporary employment is associated with a form of labour management that aims to minimise work force use, which reduces the hiring period (Banyuls and Recio 2015). 

Resorting to temporary employment depends very much on the activity sector. It responds in part to specific situations such as, for example, the high seasonality of tourism or agriculture. In the year 2017, seasonality varied between 60.7% in agriculture, 40.9% in construction, 38% in hotel and catering, and 7.6% in the finance sector, 15% in energy, water and sanitation and 17% in information and communications. In 2008, temporary employment represented 27.7% of salaried employment, this in part being due to the enormous share of temporary employment in construction (in this sector it reached 45% and alone it represented 16.3% of all temporary employment). In 2017, the percentage returned to 26.7% despite construction being a long way from recovering its previous share. This recovery of the high level of temporality appears contradictory with one of the objectives used to justify the latest labour reforms: promoting stable employment by giving more power to companies to set employment conditions and wages.

In summary, salary devaluation has arisen due to the combination of different elements: salary moderation and the stagnation of collective bargaining, the growth in jobs with lower wages (temporary and part-time), the higher proportion of jobs in low-salary sectors (hotel and catering) or the renewal of workforces in which new employees earn less than old ones. 

To interpret these changes in wages correctly, we must take into account the fact that the Survey on Salary Structure tends towards attenuation of real changes, not only because, as indicated, some low-salary sectors are not included in the survey, but also because it is difficult for the salaries of senior management to be included properly. It is also possible that the annual incomes of temporary workers are overvalued, as adequate consideration is not taken of the period of time in which they are employed. According to the data produced by Spain’s Tax Agency, nearly 30% of people who stated they had received wage incomes over the course of the year did not reach incomes equivalent to the interprofessional guaranteed minimum annual wage, a situation explained by the fact that many people are only employed for exceedingly short periods of time.

The increase of salary inequalities is not distributed evenly, but rather has affected especially certain groups of workers. One of these groups is women and the other is foreign workers.

As for gender inequalities, these are significant and persistent. Women are in a situation of inferiority with respect to men in practically every aspect of employment conditions taken into account: higher unemployment, temporary employment and part-time employment rates and a higher presence at lower levels of the labour hierarchy in professionalised sectors such as teaching and healthcare. 

One of the simplest ways of measuring this inequality is to calculate the salary gap; in other words, the average salary difference that exists between the overall set of men and women. This may be a good indicator of inequality, although its interpretation is subject to various readings. The most restrictive consists of considering that there is only gender discrimination when a man and a woman with the same education levels, type of contract, activity sector and employment position earn different salaries. In this case, inequality falls to a level close to 10%. 

The most extensive interpretation considers that the total difference is a better representation of real inequality because gender discrimination not only includes salary discrimination but also a different allocation of men and women in different jobs and even an under-evaluation of traditionally female jobs. The latter is the value that the salary gap considers.  

Thus, women earn less than men for a diverse set of reasons: they occupy many part-time jobs; they are over-represented in low-salary sectors; they occupy lower positions in the employment hierarchy; they suffer discrimination in questions of promotion; many female jobs have a low level of social recognition etc. Furthermore, and this complicates the situation, a large part of the posts that women most frequently occupy are concentrated into two segments of the labour market: professional jobs – especially in public services such as healthcare or education – and manual jobs in services.

The salary gap covers all of these circumstances and, therefore, cannot be interpreted in any simple way. For example, if employment grows in professional sectors preferably occupied by women, it is possible that the size of the gap will tend to diminish, although discrimination due to gender reasons may persist. Therefore, a variation of a few points in the value of the gap cannot be taken alone as a solid base for interpreting that inequalities are increasing or decreasing.

As we can observe in graph 4, the salary gender gap stands at above 20%, a level that is sufficiently high to indicate that we are facing significant inequality. It grew until 2013 and then underwent moderation, although this might be more a reflection of the recovery of employment (especially in traditionally feminised sectors) than of any significant change in gender inequalities.

Gender inequalities are important but not the only inequalities. Other wage differences can also hide discrimination that affects different groups of people. Nationality is one of the possible causes, as habitually a large part of the working population of foreign origin (with the notable exception of the employees of multinational companies and elite professionals) is usually concentrated into lower paid activities. This is thus not only because they are the last to reach the labour market and, due to their economic precarity and often a lack of qualifications, they are more prone to accepting less highly paid jobs, but also because migratory and other related policies (for example, the recognition of professional qualifications) push them towards marginal labour markets.

What we can say with the statistical evidence available is that this wage gap associated with nationality of origin does exist and is especially important for people originating from Latin America and the rest of the world (Asia and Africa). The wage gap with these immigrants and, especially, with those originating from Asia and Africa, has also tended to grow with the crisis, a tendency that is exactly the opposite to that experienced by European Union immigrants (graph 5). 

The crisis has led to an important increase in inequalities in income and the proportion of the population living at risk of poverty. This situation is the result of increased unemployment, cuts in social policies and a fall in wage incomes. 

In our study we have focused on wage inequalities and employment conditions. Instead of a fall in the overall set of salaried incomes that would have equally affected the salaried population and kept the already significant wage inequalities stable, what we confirmed in contrast is that these inequalities have increased, above all because what has worsened most are the wages of sectors that were already on the lower rungs of salary distribution. 

This worsening has been the result of a combination of changes in labour legislation and the growth in temporary and part-time jobs. Although this increase in wage inequalities has a lesser impact than the other factors cited, it compounds and worsens the situation, it shows inequality by gender and origin, and is in part responsible for the growth in the “working poor”, a relatively new phenomenon in Spain. 

Inequalities have become one of the central questions in economic and social debate in recent years (Piketty 2014, Stiglitz 2015, Milanovic 2016, Grimshaw et al.,2017). Their effects influence such diverse aspects of life as coexistence, education, health, et cetera. And, above all, extreme inequality subjects a part of the population to intolerable standards of living. The fact that the very functioning of the labour market causes a part of the salaried population to sink into situations of poverty and precarity constitutes a problem of considerable magnitude, as it has many negative collateral effects, such as discouraging people with regard to seeking employment, growth of the informal and illegal economy, and increases in the demand for social services.  

BANYULS, J., and A. RECIO (2015): "Gestión empresarial y dinámica laboral", Ekonomiaz. Revista Vasca de Economía, 87.

GRIMSHAW, D., C. FAGAN, G. HEBSON and I. TAVORA (2017): Making work more equal. A new labour market segmentation approach, Manchester: Manchester University Press.

MILANOVIC, B. (2017): Desigualdad mundial: un nuevo enfoque para la era de la globalización, México: Fondo de Cultura Económica.

PÉREZ AMORÓS, F., E. ROJO TORRECILLA and H. YSÁS MOLINERO (2016): Balance de la reforma laboral, Albacete: Bomarzo. 

PIKETTY, T. (2014): El capital en el siglo XXI, Madrid: Fondo de Cultura Económica.

STIGLITZ, J. (2017): La gran brecha: qué hacer con las sociedades desiguales, Barcelona: Debolsillo.