Living Standards Measure
The Living Standards Measure (LSM) is a multivariate segmentation tool employed in South Africa to classify households by their material living standards and approximate disposable income, primarily through indicators of ownership such as durable goods, appliances, vehicles, and access to utilities like electricity and piped water.[1] Developed by the South African Audience Research Foundation (SAARF) in the late 1980s and first incorporated into the SAARF All Media and Products Survey (AMPS) in 1990, it segments the population into ten ordinal groups from LSM 1 (representing the lowest standards, often rural households with minimal assets) to LSM 10 (encompassing affluent urban households with extensive possessions).[2] This asset-based approach avoids direct reliance on self-reported income, which can be unreliable due to underreporting or variability, instead using observable proxies to gauge wealth accumulation and consumption capacity.[3] Widely adopted in marketing, advertising, and consumer research since its inception with an initial set of 13 variables in 1989, the LSM has facilitated targeted strategies by transcending traditional demographics like race or income brackets, enabling firms to align products with household capabilities—such as higher penetration of branded goods in LSM 8-10 segments.[1] By 2001, SAARF refined it into a "Universal LSM" with ten standardized categories applicable across surveys, reflecting shifts in ownership patterns like rising appliance prevalence post-apartheid.[4] Its empirical foundation has tracked national progress, for instance, documenting a decline in the LSM 1 share from nearly 20% in 1994 to about 5% by 2001 amid economic liberalization and electrification drives.[5] However, the LSM has drawn scrutiny for potentially overstating living standards in unequal contexts by emphasizing assets over liabilities, ongoing poverty, or service quality, as households may acquire goods via debt without sustainable income—exacerbating South Africa's high Gini coefficient despite apparent upward mobility in ownership metrics.[6] This has prompted transitions to successors like the Socio-Economic Measure (SEM), introduced around 2017, which incorporates attitudinal and behavioral data alongside assets for a more nuanced proxy of prosperity, though LSM remains influential in legacy datasets and export-oriented market analyses.[7]Overview and Definition
Purpose and Conceptual Foundation
The Living Standards Measure (LSM) was established by the South African Advertising Research Foundation (SAARF) as a multivariate segmentation index to categorize South African households by socioeconomic status, primarily for marketing, advertising, and consumer research applications. Introduced in the late 1980s, its core purpose is to divide the population into ten hierarchical groups—from LSM 1, representing the lowest living standards, to LSM 10, the highest—facilitating targeted strategies in media planning, product distribution, and market analysis where direct income surveys prove costly, inconsistent, or evasive due to respondent reluctance.[3][8] This tool addresses empirical gaps in traditional metrics by leveraging proxy data that correlates strongly with disposable income and purchasing behavior, enabling businesses to allocate resources efficiently across diverse demographic segments.[9] At its conceptual core, the LSM rests on the observable correlation between material possessions, infrastructure access, and sustained economic capacity, positing that ownership of durable assets and services reflects long-term consumption patterns rather than transient income flows. Drawing from household survey data in the All Media and Products Survey (AMPS), it employs statistical techniques such as discriminant analysis to weight and combine variables—including appliance ownership (e.g., refrigerators, televisions), vehicle possession, dwelling type, electrification, and urbanization degree—into a composite score for classification.[3] This foundation prioritizes causal indicators of wealth accumulation and lifestyle durability over volatile self-reported earnings, which often understate or overstate actual living standards due to seasonal employment, informal economies, or reporting biases prevalent in South Africa's context.[3] By 2004, the index incorporated up to 29 such variables, refined periodically to maintain relevance amid technological and infrastructural shifts.[3] The measure's design underscores a pragmatic realism in economic proxying: asset-based indicators capture multidimensional material welfare—encompassing not just current affordability but historical financial stability—outperforming income alone in predictive power for consumer markets.[3] It deliberately omits non-material dimensions like health or education to focus on commercially actionable traits, ensuring applicability in a developing economy marked by inequality and uneven data quality.[3] This approach has validated its utility through consistent alignment with observed consumption disparities, though its market-centric lens limits broader welfare assessments.[8]Key Characteristics
The Living Standards Measure (LSM) is a multivariate segmentation tool that classifies households or individuals into discrete groups based on observable indicators of living standards, eschewing direct income data in favor of more stable proxies for wealth and consumption capacity. Developed as a market research instrument, it aggregates scores from a predefined set of variables to form a composite index, typically dividing populations into 8 to 10 hierarchical segments, with higher segments denoting elevated standards.[8][3] Central to its methodology are indicators centered on asset ownership and infrastructural access, including possession of durable consumer goods such as televisions, refrigerators, automobiles, and major appliances; housing characteristics like construction type (e.g., brick vs. informal structures) and availability of electricity or piped water; and contextual factors such as degree of urbanization.[8][3] In the South African implementation, managed by the South African Advertising Research Foundation (SAARF), these elements yield 10 segments from LSM 1 (minimal assets, rural or informal settings) to LSM 10 (abundant durables, urban formal housing).[8] LSM's design prioritizes empirical verifiability over self-reported earnings, mitigating inaccuracies from income underdeclaration prevalent in surveys from emerging markets, and demonstrates stronger correlations with purchasing behavior, media usage, and lifestyle patterns.[3] This asset-focused approach enables reliable consumer profiling for advertising and product targeting, as ownership reflects cumulative economic status more enduringly than volatile income flows.[8] While adaptable across regions like India and South Africa, variations in indicator selection ensure cultural relevance without altering the core emphasis on tangible living conditions.[3]Historical Development
Origins in the 1980s
The Living Standards Measure (LSM) originated from efforts by the South African Advertising Research Foundation (SAARF), established in 1974 to standardize media and market research, to create a reliable segmentation tool amid the limitations of income data and demographic proxies in apartheid-era South Africa. In the late 1980s, SAARF developed LSM as a multivariate index focusing on household ownership of durable goods and services, such as electricity, refrigerators, and hot water systems, to gauge purchasing power and living standards without relying on self-reported income, which was often inaccurate due to informal economies, subsidies, and political sensitivities. This approach drew from earlier work by Unilever researcher Eddie Schulze, who in the 1980s pioneered a classification system based on asset ownership to predict consumer behavior, filling a gap left by racially charged or unreliable traditional metrics.[4][10] The initial LSM model was derived from statistical analysis of 71 potential variables collected through SAARF's All Media and Products Survey (AMPS), which were reduced to 13 key indicators exhibiting the strongest discriminatory power in segmenting households into socioeconomic groups. These variables emphasized tangible markers of modernization and affluence, reflecting causal links between asset accumulation and disposable income capacity, rather than subjective or volatile financial declarations. The tool was first incorporated into SAARF's 1989/1990 AMPS reports, dividing the population into eight LSM categories (later expanded), with higher groups indicating greater access to amenities and media.[3][2] LSM's emergence addressed the need for a neutral, empirically grounded alternative to race-based or income-centric segmentation, which were fraught with measurement errors and ideological biases in a society marked by enforced segregation and unequal resource distribution. By prioritizing observable, convergent consumer behaviors tied to living standards, it enabled marketers to target audiences based on actual lifestyle patterns, bypassing the distortions of apartheid policies that suppressed formal income reporting among non-white populations. Early adoption by SAARF validated its utility in correlating with media consumption and product ownership, establishing it as a cornerstone for South African research despite critiques of its insensitivity to rapid socioeconomic shifts.[11][12]Evolution Through the 1990s and 2000s
During the 1990s, the Living Standards Measure underwent periodic refinements to better capture shifts in household asset ownership and access to services amid South Africa's socioeconomic transitions. In 1993, the original set of 13 variables from the 1989/90 iteration was adjusted by removing indicators for "no VCR set" and "no tumble dryer" while adding "microwave oven" and "metropolitan dweller" to reflect emerging consumer durables and urban-rural distinctions.[3] By 1995, further modifications included the removal of "sewing machine" and "metropolitan dweller," alongside the addition of 10 new variables such as "flushed toilet," "hot running water," and income-related proxies like "no financial services used" and "no credit card," expanding the tool's scope to incorporate sanitation, utilities, and financial access as proxies for living standards.[3] These updates were driven by data from the All Media and Products Survey (AMPS), with variable selection informed by statistical techniques like discriminant analysis to maximize differentiation across LSM groups while maintaining relevance to evolving household possessions.[3] The revisions aimed to address criticisms of outdated indicators, ensuring the measure remained a reliable segmentation tool for marketing despite rapid changes in technology and infrastructure post-1994.[3] Entering the 2000s, the LSM continued to adapt to technological advancements and broader data availability. In 2000, five variables were added—including "built-in kitchen sink," "car/sedan," and "electric stove"—while four were removed, such as "rural dweller" and certain redundant ownership negatives, refining the index to emphasize modern appliances and mobility.[3] This period marked a shift toward incorporating user feedback and more comprehensive AMPS datasets, enhancing predictive power for consumer behavior.[3] A significant advancement occurred with the launch of the SAARF Universal Living Standards Measure (SU-LSM) around 2004–2005, which expanded to 29 variables to provide greater granularity and universality across surveys.[3] Key additions included "DVD player," "one cell phone in household," "house/cluster house/town house," and the reinstatement of "metropolitan dweller" and "sewing machine," reflecting the proliferation of digital media, mobile technology, and housing types in the early 2000s.[3] The SU-LSM methodology emphasized continuous development through empirical validation, positioning it as a more robust alternative to income-based metrics by tracking asset accumulation as a stable indicator of living standards.[3]Recent Updates Post-2010
Since 2010, the Living Standards Measure (LSM) developed by Colmar Brunton has maintained its foundational methodology of classifying New Zealand households into 10 socio-economic segments based on ownership of durable goods, access to utilities and transport, and leisure activities, without publicly documented structural revisions to the core framework. Annual or periodic surveys by Colmar Brunton (subsequently integrated into Kantar following acquisition) continue to refresh the underlying data, allowing segments to adapt to empirical shifts such as rising household internet penetration—from 72% in 2010 to over 90% by 2020—and smartphone adoption, which correlate strongly with higher LSM groups (e.g., LSM 9-10 households exhibiting near-universal digital access by mid-decade). These data-driven adjustments ensure the tool's utility in media audience profiling, where LSM remains a standard covariate in ratings systems like those managed by NZBARB, though it has faced calls for greater integration with behavioral digital metrics amid streaming's growth. No peer-reviewed studies or official releases indicate fundamental changes to variable selection or weighting post-2010, underscoring the measure's robustness but also potential limitations in capturing non-material or transient economic factors like gig economy participation.Methodology and Classification
Core Variables and Indicators
The Living Standards Measure (LSM) relies on a selection of proxy indicators for household wealth and access to amenities, deliberately excluding direct income metrics to mitigate reporting inaccuracies and focus on observable living conditions. Developed by the South African Advertising Research Foundation (SAARF) in collaboration with partners like ACNielsen, the system initially drew from 71 characteristics in 1989, narrowing to 13 key variables via discriminant analysis to maximize segmentation power across living standards.[13][14] By the 2000s, this expanded to 29 variables, periodically updated to reflect technological and socioeconomic shifts, such as replacing outdated items like sewing machines with stronger predictors like satellite television access.[15][16] These indicators, sourced from surveys like the All Media and Products Survey (AMPS), emphasize durable goods ownership, utility access, and dwelling characteristics, enabling statistical classification into deciles from LSM 1 (lowest) to LSM 10 (highest).[3] Core variables cluster into categories reflecting infrastructure, consumer durables, and lifestyle proxies. Access to services includes hot running water, electricity for lighting and cooking, piped water inside the dwelling, flush toilets (inside or outside the home but connected to sewage), and paved roads to the property.[17] Ownership indicators cover essential appliances like refrigerators/freezers, televisions, and radios; labor-saving devices such as washing machines, vacuum cleaners, and microwaves; and advanced items including personal computers, hi-fi systems, and DVD players.[18] Transportation and communication variables feature car or bakkie (pickup truck) ownership, fixed landline telephones, cellular phones, and satellite decoder ownership.[8] Additional factors account for dwelling type (e.g., brick structure vs. informal shack), employment of domestic workers, and urbanization degree, with higher LSM groups correlating to formal housing and full service access.[19]| Category | Example Indicators | Discriminatory Role |
|---|---|---|
| Utilities & Sanitation | Electricity, hot running water, flush toilet with sewage, piped water in dwelling | Proxy for basic infrastructure reliability; near-universal in LSM 8-10, rare in LSM 1-3.[17] |
| Kitchen & Laundry Appliances | Fridge/freezer, washing machine, microwave | Indicate capacity for food storage and convenience; fridge ownership exceeds 90% in LSM 5+, under 20% in LSM 1.[15] |
| Entertainment & Tech | TV, DVD/VCR, cellphone, satellite dish | Reflect media access and modernity; cellphone penetration high across groups but satellite TV concentrated in LSM 7+.[16] |
| Transport & Other | Car ownership, domestic worker employed | Signal mobility and affluence; car ownership under 5% in LSM 1-4, over 80% in LSM 10.[8] |