This dissertation focuses on the operationalisation of the concept of poverty. It analyses the capacity of the consensual deprivation indicator and reference budgets (RBs) in establishing the minimum acceptable living standard in Finland. Based on this examination, this dissertation introduces alternative indicators based on the consensual deprivation indicator and RBs.

Alleviating poverty is considered to be the primary task of a welfare state. In wealthy countries, poverty is typically understood as relative poverty. Relative poverty is typically defined as individuals being unable to participate in the minimum acceptable living standard of their society due to a lack of resources. Therefore, relative poverty involves the lack of social efficiency.

Numerous poverty indicators have been developed for operationalising the concept of relative poverty. There are different factors that separate poverty indicators. One relates to whether poverty is measured through resources—such as household income—or living standards; these indicators are described as indirect and direct indicators, respectively. The second difference relates to the source of expertise. The source of expertise refers to whose decision the operationalisation of poverty is based on. This dissertation focuses on public-led and expert-led poverty indicators. For public-led indicators, the minimum acceptable living standard is based on the assessment of the public. Common to the public-led indicators is that some sort of consensus is assumed about the minimum acceptable living standard. In expert-led indicators, the decisions about the minimum acceptable living standard are made by experts, such as researchers.

This dissertation presents six different poverty indicators and examines their capacity to establish a minimum acceptable living standard. Frequently, poverty is measured using the at-risk-of-poverty (AROP) indicator, in which the poverty threshold is set at 60% of the equivalised national median income. The poverty threshold is set based on expert opinion. Even though this indicator is widely used, it has several acknowledged problems; thus, deprivation indicators are being increasingly used to measure poverty. Deprivation indicators focus on the lack of consumption items. There are different deprivation indicators with differences in how the minimum acceptable living standard is established. Relative deprivation examines necessities, which are defined by experts; however, it makes use of income to derive the poverty threshold. In this sense, it can be seen as an income-based indicator. For the consensual deprivation indicator, the necessities are defined by the public and the poverty threshold is understood as an enforced lack of these necessities. The material deprivation indicator has similarities with the consensual deprivation indicator; however, the minimum acceptable living standard is considered to be supranational, rather than national.

Additionally, there are RBs, which are considered to be the oldest indicator of living standards. RBs are priced baskets of goods that can be constructed to present the minimum acceptable living standard. This is done by taking into account the institutional and social context of the given society. However, there is not a shared methodology for constructing such RBs. RBs can be constructed either based on expert knowledge or the information provided by focus groups.

The dissertation comprises four sub-studies. The first sub-study examines whether a consensus can be established using the consensual deprivation indicator. This is analysed using two criteria established in earlier research. First, people should base their assessment on public evaluations, rather than private judgments, of the necessity for certain items. Secondly, people should agree on what the necessities are. Typically, this is analysed at the group level; however, in this sub-study the analysis takes place between individuals. The results indicated that the consensual deprivation indicator was problematic in terms of establishing the minimum acceptable living standard. Further, people’s views on the necessities were shaped by their own preferences. Additionally, consensus regarding the necessities was modest at best.

The second sub-study examines whether two RBs produce similar results. Two Finnish RBs provide an excellent opportunity, as they target the same living standard and are constructed for the same period and population but are based on different methodologies. This sub-study provides explanations of why differences between RBs occur. Building on the findings of the first and second sub-studies, this dissertation introduces alternative poverty indicators. The results indicate that the two RBs produce different estimates about the resources needed for the minimum acceptable living standard. Notably, the differences involved mainly the costs for housing and mobility. The differences in the estimates were traced back to differences in information bases, selection criteria and pricing of the items.

The third sub-study examines poverty in Finland using a poverty indicator based on RBs. The results regarding the prevalence and concentration of poverty of the RB indicator were contrasted with those of the AROP indicator. The results indicated that the poverty rate was lower with the RB indicator, compared with the AROP indicator. Additionally, the risk groups of poverty were somewhat different. In particular, this concerns the elderly, as the poverty rates for this group were considerably lower when using the RB-based indicator.

In the fourth sub-study, several different weighting approaches are used for analysing consensual deprivation. Typically, using the consensual deprivation approach, the items are unweighted, thereby attributing the same importance to all items. This sub-study argues that by using the weighted approach, some of the problems regarding typical consensual deprivation indicators can be avoided. The results indicate that weighting the items changes the prevalence of material deprivation but only modestly.

In sum, this dissertation illustrates that the consensual deprivation indicator does not lead to a unified view of the minimum acceptable living standard. Similarly, different RBs produce different estimates of the minimum acceptable living standard. Nonetheless, RBs can address some of the problems that are embedded in the AROP indicator. However, further methodological work is needed. In a similar vein, the weighted consensual deprivation indicator reflects more accurately the minimum acceptable living standard in Finland, compared with the unweighted approach. These alternative poverty indicators have the potential to be used in poverty research to improve the accuracy of poverty measurement.