Main page




by Valentino Piana (2004)



1. Significance
2. Social groups defined along income and non-income axes
3. Core vs. periphery
4. The boundaries of social groups
5. Self-consciousness of social groups
6. Formal models
7. Data
  8. Related papers  


Throughout the world, societies are split into distinct, albeit sometimes overlapping, social groups.

While self-definition - and self-defined belonging to groups - widely varies over time and situations for even the same individual, there are several indicators and features along which the researcher can build clusters of people.

Indeed, the atomistic idea according to which everyone feels to be dissimilar from anybody else is a realistic picture only of very extreme social conditions. Networks of people linked by communication and emotional lines are very common, while their collapse into monistic units of disconnected individuals is more the result of crashes in the social texture than the standard state.

This is very important for economics since, for instance, social groups are important for consumption as visible behaviour can become a standard with which to compare your owns. Peer groups are often the reference for choosing whether a certain good is a "must have", engendering an imitation dynamics.

In this vein, another name for "social group" is "market segment", seen from the point of view of a seller or a mass medium, to which a differentiated product can be addressed.

Social groups defined along income and non-income axes

A basic axis of social differentiation is personal income. A three-category distinction among the poor, the middle class and the rich, as the one we proposed here, can be too rough to capture all the intermediate nuances but still help characterize different societies.

The unit of analysis can be - but not need to be only - the individual. A second element should be kept into account when judging the individual's position: the size and composition of the household to which he or she belongs. Larger families pay less per-person for certain shared expenditure (like housing) and they can have more than one income-bearer. If not, a large family with only one income-bearer would be much poorer than a one-component family.

A third consideration is linked to the stability of income over time, with people having larger fluctuations in income belonging to different social groups than people reaching systematically and without renegotiation similar levels of income.

Another crucial axis of social differentiation is the ownership of assets. Families that own their home - having already paid back any financial instruments to buy it - can devote more money to active savings and consumption than families with the same level of income but that have to pay the rent. The ownership of durable goods is an important status element aiding different aspects of life. Financial assets (as Treasury bonds, shares, controlling majorities in firms,...) can constitute elements of common material interests for certain social groups.

Employment is an extremely important element in defining identities and common interests, common languages and values. Equally important are culture and levels of education.

Core vs. periphery

Economists often ignore many of those axes and are tempted to treat each of them as uncorrelated to the others.

But sociology has convincingly shown that these axes tend to covariate, so that rich people tend to be more educated, to have top jobs, to participate more to social life and decision-making.

Thus one should distinguish a core of society (people with more than certain thresholds of income, education, ... and strong connections through social communication channels) - which actively takes part to public life - in contrast to social periphery, a more or less marginal layer of people (poor, emarginated, isolated) [1].

Social cohesion is what keep society united, avoiding the polarization of core vs. periphery. To measure social cohesion, one should look at how deep is the gap between the so-called "have's" and "have not's" (as in the case of "digital divide", the access to clean water, or the softer concept of "employability") and how fast is this gap filled through a diffusion dynamics.

Electoral participation is much higher in the core than in the periphery. This explains - in part - why, even in quite polarized society, democratic election can be won by parties making the interests of the small minority of rich.

More in general, social groups may shift to support the same party or policy option as a block, with politicians trying to involve influencial members of the social group into their campaigns and apparatus to reach out to the overall social group. They may develop a specific language and key words decoded by the members as attractive, including derogative references to opposed social groups.

Electoral pyramids of people in connection voting as a block are immersed in a sea of un-connected or pluri-connected people who vote independently or unconsistently. Political tactics can leverage on pyramids or be conducted through direct appeal to voters.

Core, periphery and semi-periphery can also characterise geographical areas and organizational structures (as firms).

The boundaries of social groups

Groups can be defined as people doing the same thing, liking the same subjects, sharing the same opinions, aiming at the same goals, using the same line of reasoning and behavioural routines. More technically, groups have common interests, share real experience, express similar way of thinking and reacting.

In a further dimension, one can look at social groups as connected networks of people. They actively exchange information and Weltanschauungen. In the presence of new events, they consult each other on the interpretation to give to them and to decide what to do.

Solidarity networks link families, relatives, friends, neighbours up to co-citizens and people even if not previously known.

The geographical boundaries of social groups are a distinct aspect to analyse, since they are partially influenced by similarities across nations as well as by the hierarchical structure of international relationships.

Self-consciousness of social groups

One can have a strong consciousness of belonging to a certain group, who has specific material and non-material common interests as well as characteristic way of developing strategies, organizing and acting. But this is not always the case, since it is the result of personal, generational, collective experience.

Usually, the researcher takes existing groups and then look at how they react to events and issues. But one can take the other way round and look how social identities arise from conflict and confrontation with "strangers". It's when you are under attack from "somebody" that you look at possible helpers - people on the same part of the barricades. This happens for instance when a state policy is hurting a social group. A recognized leadership and an organizational hierarchy help a lot in transforming a latent identity into a co-ordinated multi-agent force.

In more "peaceful" situations, an individual may belong to more than one group and the axis - along which he feels to be defined - can be changed by mass media pressure as well as by specific events.

In a macro-perspective, "totalizing identities" as ethnicity, language, religion can reduce the subjective importance of other social boundaries.

In a micro-perspective, the consciousness of collective identity can arise from current participation to social activities and events, especially when they have a characteristic mass dimension: large factories, manifestations, church rites, stadiums, music concerts.

Formal models

Income distribution and the diffusion of an innovation in social groups

Waves in consumption with interdependence among consumers


Identify social groups along income and non-income axes in this micro-data set

Readership by job

Consumers data (income, preferences for performance and comfort, decision rules)

Daily time spent on different activities by employment status, sex, and age

Related papers

On the identification of the “middle class” (Sept. 2011)
Income polarization should leave room to a much stronger and wider middle class. But what is "middle class" nowadays? This paper explores theoretical and practical definitions that can be the sound base for new policies.


[1] A similar situation is found among nations, as explained in this paper where we provided a more formal definition of "core", "periphery", and "semiperiphery".

Key concepts
  Labour market