Low-wage employment Updated

Who are the low-paid?

In general, wages are considered low if they are less than two-thirds of the national median or the national mean of gross hourly wages. This definition of low wages, which has been endorsed by international organizations like the OECD and the EU, avoids the difficulty of defining an absolute level of low pay and facilitates international comparisons. Although other definitions of low pay are used—for instance, the lowest three deciles, or half the mean of the wage distribution—the ranking of countries in terms of low-pay incidence or persistence and the insights from empirical studies seem to be robust to using alternative definitions of low pay [5].

A number of studies across various countries have tried to identify individuals who are low-paid as well as their defining characteristics. Data from the European Community Household Panel (ECHP) 2001 indicate that, among all employees, the incidence of low pay is twice as high for female workers as for male workers [6]. Low-pay incidence is also particularly high for young workers, for those in temporary contracts, and for agricultural workers. The probability of being low-paid tends to decrease with a worker's qualification level (Figure 2), but it is notable that (in some countries) significant shares of skilled workers are also engaged in low-wage employment. These findings are confirmed by other studies for the EU [5] and for a number of countries like Denmark, the Netherlands, Italy, and Spain [7]; Germany [8]; the UK [9]; and Australia [3].

Upward mobility of low-paid workers

Having identified the main characteristics of low-paid workers, it is important to know which factors affect the dynamics of low-pay employment; in other words, do workers typically remain in low-paid jobs or are they able to move up to better-paid jobs? Studies for a number of countries show that workers in low-paid jobs are quite likely to remain in low-wage employment from one year to the next. For instance, data from the ECHP indicate that, across the EU-15 countries, about half of those workers who were low-paid in 2000 were still low-paid in 2001, whereas 31% managed to obtain wages above the low-wage threshold and almost 18% moved into non-employment [6]. Taking into account a longer time perspective of five years, 30% of the low-paid workers from 1994 remained in this state, while 43% managed to increase their wages above the low-pay threshold (and 27% ended up in non-employment). Using the same set of data, a comparative analysis of male workers in 12 EU countries during the period 1994−2001 finds that the probability of a worker remaining in low-paid employment over two successive years varied from 49% (in Spain) to 70% (in the Netherlands) [5]. The authors of this study point out that, in general, countries with a relatively large share of low-paid workers also seem to be those countries where it is most difficult to move out of low-wage employment.

In the US, over the period 1968–2014 more than half of all full-time workers entering low-wage employment moved above the two-thirds of the median threshold within four years. However, mobility out of low-wage employment has declined since the late 1990s, in particular for manual workers [10].

Detailed empirical analyses for various countries indicate that a number of factors influence labor market transitions from low to higher pay. These include individual characteristics as well as sector-specific, occupational, and establishment characteristics. Furthermore, the overall state of the labor market and labor market policies (such as training and public employment policies) may play a role too.

Concerning individual worker characteristics, not only is low-wage employment more prevalent among female and low-skilled workers, but the chances that these groups are able to escape from low-paid jobs are also generally more limited [6], [7], [8]. The same applies for non-white workers in the US [10]. In contrast, younger workers at the beginning of their careers exhibit a higher probability of moving up the wage ladder than older workers [6], [8].

The characteristics of the employing firms are also significant for low-wage earners’ chances of achieving higher-paid jobs. In Germany, chances are better in larger industrial plants, which more frequently offer training and other possibilities for workers to accumulate human capital than smaller plants [8]. Moreover, plants with a high share of low-paid workers appear to be dead ends for individual low-paid workers, since the chance of upward wage mobility in these plants is significantly lower [8]. Sectoral affiliation also seems to play a role; workers in agriculture are particularly unlikely to leave their low-wage jobs, whereas workers in the public sector seem to have better chances of securing higher-paid jobs in the future [6], [10].

Interestingly, there is no clear-cut evidence on how overall economic conditions and the state of the labor market affect transitions from low-paid to higher-paid employment. Having said that, a study for the EU-15 countries reports a small dampening effect of higher unemployment rates on transitions from low- to higher-paid jobs between two successive years (but no effect is seen on transitions over three years) [6]; additionally, an empirical investigation for Australia suggests that transitions between various labor force and earnings states are lower under weak than under strong economic conditions [3]. With respect to the role of labor market policies and firms’ human resource management policies, transitions from low to higher pay are more likely for those workers who have received training or vocational courses [5], and in particular for those who get on-the-job training [7].

Stepping stone and scarring effects of low-paid jobs

When interpreting the evidence on pay transitions, it is important to distinguish between the varied characteristics of workers that predispose them to low-paid jobs and the direct causal effects of low-paid employment on the probability of being low-paid, high-paid, or unemployed in the future. As discussed above, accepting a low-paid job may induce scarring effects by sending negative signals to prospective employers, by being associated with a deterioration of human capital, or by reducing workers’ search intensity. It is thus important to find out, for an unemployed individual, whether it is better to take up a low-paid job or to remain unemployed and wait for a better job offer at some later point.

By estimating models that account for the fact that people who accept low-paid work are a selected subgroup (mostly based on the analytical tools and concepts developed in [11]), some empirical studies have tried to assess if low pay during a particular period in a worker's career path has a genuine effect on his/her likelihood of being in certain labor market states in the future, even after accounting for worker characteristics that predispose them to low pay. A number of studies find that having been employed in a low-paid job does indeed seem to have a genuine effect on the probability of being low-paid, high-paid, or unemployed in the future. For instance, low-pay state dependence is found in all 12 EU countries examined by a comparative study of male workers, even after individual worker difference is accounted for [5]. State dependence is also found in two studies of workers in the UK that apply various methods of estimating the transition probabilities into and out of low-paid employment [9], [12]. Another study about British men shows that low-paid jobs have almost as large an adverse effect as unemployment on future employment prospects [13], leading to the conclusion that low-wage jobs are a conduit to recurring unemployment in the UK.

In contrast, more recent evidence from the UK [12] and from other countries suggests that, although substantial scarring effects from unemployment exist, low-paid jobs tend to improve workers’ employment and income prospects compared to being non-employed. Two Australian studies make clear that both state dependence and stepping-stone effects of low-wage employment can be present [3], [14]: workers in low-paid jobs are more likely to be in low-wage employment in the future compared with individuals who are not in the labor force, are unemployed, or are in higher-paid jobs. However, low-paid workers are also more likely to move into higher pay in the future than those who are unemployed or those not in the labor force. For Germany, two studies show that the chances of obtaining a high-paid job by accepting low-paid work improve most strongly for those who have experienced longer periods of unemployment and for those who are less-skilled [15]. On the other hand, for individuals with medium or high qualification levels, the risk of non-employment in the future is not significantly lowered by taking up a low-paid job instead of remaining unemployed [2]. The latter result stands in contrast to an Australian study that reports less severe penalties from low-paid employment for better-educated workers [14].

Focusing on women, who constitute a disproportionate share of low-wage earners, another study for (western) Germany finds genuine state dependence in low-wage employment: having a low-paid rather than a high-paid job decreases the probability of being highly-paid in the future, and this negative effect is particularly strong in part-time jobs. Only for part-time workers, low-wage employment also increases the risk of being unemployed in the next period. With respect to future wage prospects, however, low-paid women are found to be significantly better off than unemployed or inactive women. The authors argue that, for women, low-wage jobs can help lead them out of unemployment and that these jobs should be prioritized over remaining unemployed while waiting for a better job [4]. An Australian study provides corroborating evidence that low-paid employment is always preferable to unemployment for women [14].

A careful interpretation of the empirical evidence across a number of countries thus allows three main conclusions. First, there is some evidence for state dependence regarding low-paid work in many countries. Being in low-wage employment today seems to have a genuine influence on the probability of being low-paid (as well as the probabilities of being high-paid or unemployed) in the future. However, since observed state dependence might not necessarily imply causality (for instance, due to selection issues, estimation problems, and lack of data), some caution with interpretation is clearly advisable. Second, although upward mobility appears to be limited, low-wage employment is not a persistent experience or a dead end for all low-wage earners. Low-wage jobs can act as stepping stones to better-paid jobs for some groups of workers. Third, empirical evidence on the existence of a low-pay no-pay cycle is mixed: compared to being highly-paid, some studies find that having a low-paid job increases the risk of being unemployed in the next period, or that this is the case for specific worker groups [4], [13], [14], whereas in other studies [3], [12] or for other demographic groups [14], this relationship is not observed.