INTRODUCTION

This is not “simply” a study of the literature regarding wage inequality in the labor mar­ket, even apart from the fact that the literature is immense. The income distribution is the focus of the present handbook and provides the ultimate rationale for considering the dispersion of wage earnings here.

However, understanding institutions in relation to wage dispersion is our overarching purpose—and a very demanding purpose in its own right. It would require a bridge too far to also try to overcome the gap and incorporate the income distribution in our approach. Instead we will take a swift look at said literature and the stylized facts of the subject, and we will do so at the start of our argument to make the best of it as a heuristic device for our ensuing discussion of wage dispersion and institutions. Thus, we hope to make a contribution on which future analysis can expand by providing a building block that can be used subsequently for constructing a unified economic theory of income distribution, a theory that is still missing (Atkinson and Bourguignon, 2000, 26). By the way, that building block itself needs to account for the very fact that neither a unified theory of earnings dispersion is available. We intend to do that by reviewing the literature on institutions and earnings distribution in a framework that may be relevant also for further use in studying the household income distribution.

Concretely, we explicitly include in our focus the distribution of annual earnings from labor as the income distribution is commonly measured and analyzed on an annual basis.^[113] This entails, first, that we study both wage rates and (annual) hours of work—which taken together make up annual earnings—as well as the dispersion of both and their inter­relationship. Thus, we aim to go beyond, e.g., Blau and Kahn (1999), who address the effects of wage-setting institutions on wage inequality as well as on employment, but for the latter restrict themselves to aggregate employment effects and ignore its dispersion over individuals and households as well as its relationship to wage dispersion. It implies that one needs to consider the role that institutions play not only in relation to the wage rate, the hours worked, and the individual probability of employment, but ultimately also in relation to the household distribution of employment—what we can call a double­edged employment perspective.

Second, we will contemplate the relevance and the effects of LMIs from this distri­bution point of view. Particularly, we will on the one hand leave aside the literature that focuses on wage dispersion in relation to the matching of workers to given jobs (e.g., Mortensen, 2005, on search, or Rosen, 1986, on compensating differentials). We also leave out the literature on other important facets of inequality such as earnings mobility or its role as a work and career incentive. On the other hand we will look—to the extent that we can—for institutions that may affect the distribution of employment over house­holds (e.g., equal treatment, working-hours nondiscrimination, child care provisions or tax measures) or the supply of hours over the year (e.g., temp agency work, temporary contracts). Thus, different institutions from the usual suspects may come into play, for example, new rules and regulations regarding part-time jobs and pay, or the “reconciliation of work and family life,” while at the same time those usual suspects will be checked for their effects in this domain. The “new” institutions will need to be con­sidered in their own right but, naturally, also in relation to the previous ones. We need to be careful, though, that the assortment of institutions under scrutiny be manageable; as in modern society labor market behavior has become so central to human existence that virtually any institution might be thought to have an effect.

In our take on the literature, we aim to be careful in considering the role of institu­tions not in isolation of the “normal” economy.

In addition, we like to stress that the time period effectively covered in the chapter is determined by the literature that we aim to address. Though that period may seem long to some as we begin our coverage at the end ofthe 1960s for certain countries, it is important to realize that the trends found may be selective. The long-run historical perspectives adopted in the top-incomes literature (Alvaredo et al., 2013) or in Atkinson’s (2008) inter­nationally comparative study ofthe earnings distribution suggest that preceding trends may diverge, sometimes radically, and might throw a different light on the mechanisms at work.

Before continuing, we mention a caveat regarding the two concepts of “dispersion” and “inequality,” which we have used indiscriminately to indicate the squeeze or stretch of a distribution. A major reason for many to pay attention to the dispersion is that a large part ofit coincides with social or economic inequality as it is commonly understood. How­ever, more precisely, the dispersion is thought to relate to a range of observations, wages, or incomes in this case, that are not all the same and therefore are unequal in a mechanical, mathematical sense of the word. Inequality, by contrast, provides a qualifier to such observations that makes them unequal in the sense of analyses providing an explanatory interpretation of the observations, either individual or aggregated. So, strictly speaking, dispersion and inequality are different concepts. Not all mechanical differences will also be inequalities from an analytical point of view, for example, differences in individual earnings that reflect differences in efforts. Conversely, not all analytical inequalities will also be mechanical differences, for example, individual earnings that are identical in spite of differences in efforts. Having said this we will continue to use the two words interchange­ably as this chapter is aimed at evaluating a set of such qualifying analyses. Note, finally, that measures of dispersion or of inequality (Gini coefficient, etc.) are identical, and are usually called measures of inequality—terms that we will also use in this chapter.

Some of the above references indicate the existence of various literature overviews that are relevant to our study of earnings inequality, which are found in the first volume of the Handbook of Income Distribution, all volumes of the Handbook of Labor Economics, and the Oxford Handbook of Economic Inequality. We will not redo these, but gratefully build on them when it is useful to do so.

2011; DiPrete, 2007; Golden and Wallerstein, 2011; Kenworthy and Pontusson, 2005; Oliver, 2008; Wallerstein, 1999). We will also allude to some of their results.

Our contribution takes the general level of inequality as its starting point but cannot escape digging below that surface. Thus, for example, we may touch upon the tails of the distribution—top incomes, (in-work) poverty—where much of the action is. However, for a deeper understanding of those tails as well as the complementing middle we refer to the treatment of polarization (Chapter 5), top incomes (Chapter 7), and in-work poverty (Chapter 23) elsewhere in this handbook. More generally, the labor market also figures as one of the multiple causes of inequality in Chapter 19. On another dimension, our con­tribution stops short of the within-household distribution (see Chapter 16) or any further analysis of gender inequality (see Chapter 12). Finally, this chapter will cover those coun­tries that have well-developed, comprehensive formal labor markets. This restricts the selection of the literature to analyses that concern the United States, Canada, Japan, Korea, Australia, New Zealand, the member states of the European Union, and some other European countries such as Iceland, Norway, and Switzerland.

18.1.1 LayOut

The layout of the chapter is as follows. First, in Section 18.2, we will briefly discuss the literature that regards the link between wage dispersion and the household income dis­tribution, considering the distributions of earnings and employment from both the indi­vidual and the household perspectives, and presenting some stylized facts. In Section 18.3 we discuss the measurement of wage inequality with some relevant data sources and pre­sent some stylized facts of wage dispersion for a selection of countries. Next, in Section 18.4, we discuss theories aimed at explaining the dispersion of wage rates and the role of institutions. Section 18.5 then addresses the role of LMIs empirically, with the help of a model that that incorporates several features advocated in the preceding sec­tions, such as a focus on earnings, i.e., the product ofwage rate and annual efforts, and that inserts as explanatory variables a number of “new institutions” related to household labor supply. In addition, we use recent internationally comparative data. Finally, we conclude in Section 18.6 by summarizing the main findings and considering issues warranting further research.

18.2.