Social, Labor, Poverty & Sustainable Development Statistics in Official Governance

Labor force statistics measure the mobilization of human resources in economic production, the standard of living of households, and the structural dynamics of employment, unemployment, and labor underutilization.

In 2013, the 19th International Conference of Labour Statisticians (ICLS), convened by the International Labour Organization (ILO), adopted a revolutionary resolution fundamentally reforming the statistical definition of work. The 19th ICLS framework established that work encompasses any human activity producing goods or services for consumption by others or for own use. It partitioned work into five mutually exclusive forms:

1. Own-Use Production Work: Producing goods or services for own final household consumption (e.g., subsistence farming, collecting firewood, fetching water, home cooking, childcare).

2. Employment Work: Work performed in exchange for pay, profit, or family gain.

3. Unpaid Trainee Work: Apprenticeships, internships, and clinical residencies performed without financial compensation.

4. Volunteer Work: Non-compulsory work performed without pay for charitable organizations or other households.

5. Other Work Activities: Unpaid community service, mandatory work by prisoners.

Crucially, only Employment Work enters the official Labor Force. The other four forms of work are recognized as productive human activity but are tracked via separate satellite accounts, preventing the dilution of economic employment metrics.

Decomposition of the Working-Age Population

The national statistical system defines the Working-Age Population ($WAP$) (typically individuals aged 15 years and older residing in private households, excluding institutional populations in prisons or military barracks). The working-age population is divided into two mutually exclusive categories:

The Labor Force ($LF$) comprises all persons who supply labor for the production of goods and services for pay or profit:

Operational Definitions

1. Employed ($E$): All persons of working age who, during a short reference period of seven days (the reference week), engaged in any activity to produce goods or provide services for pay or profit for at least one hour. The one-hour criterion ensures that all forms of precarious, casual, part-time, and informal work are captured. The employed category also includes persons temporarily absent from a job (due to illness, annual leave, maternity leave, or strike).

2. Unemployed ($U$): All persons of working age who simultaneously satisfy three strict criteria during the reference period:

• Without Work: Not employed (did not work even one hour for pay or profit).
• Currently Available: Available to start work within a specified short period (typically within the reference week or following two weeks).
• Actively Seeking Work: Took specific, verifiable steps in a specified recent period (past four weeks) to seek paid employment or self-employment (e.g., applying to job postings, submitting CVs, registering at employment exchanges, placing advertisements, applying for business loans).

3. Outside the Labor Force ($OLF$): All working-age individuals who are neither employed nor unemployed (full-time students, retired persons, homemakers engaged exclusively in unpaid domestic duties, persons with severe disabilities, and discouraged job-seekers).

Core Headline Indicators

1. Labor Force Participation Rate ($LFPR$):

2. Employment-to-Population Ratio ($EPR$):

3. Unemployment Rate ($UR$):

Notice that the Unemployment Rate is calculated as a percentage of the Labor Force, not of the total population!

In developing and low-income economies (such as Bangladesh, India, or Kenya), the headline Unemployment Rate is frequently surprisingly low (e.g., $3\%$ to $5\%$). This paradox occurs because in the absence of comprehensive universal unemployment insurance, impoverished individuals cannot afford to remain idle while actively searching for formal jobs. They must survive by engaging in precarious informal activities—selling tea, rickshaw pulling, day labor—which classifies them as "employed" under the one-hour rule.

To provide an honest, multidimensional diagnostic of labor distress, the ILO 19th ICLS resolution formulated four composite Measures of Labor Underutilization (LU1 to LU4):

1. Time-Related Underemployment ($TRU$)

Persons in employment who wanted to work additional hours, whose working hours were below a designated threshold (typically 35 or 40 hours per week), and who were available to work additional hours.

2. The Potential Labor Force ($PLF$)

Working-age individuals outside the labor force who exhibit an attachment to the labor market, consisting of:

• Unavailable Jobseekers: Persons who actively searched for work but were not immediately available to start.
• Available Potential Jobseekers (Discouraged Workers): Persons who want to work and are immediately available, but did not actively seek work during the past four weeks because they believe no suitable jobs exist, they lack skills, or they gave up searching.

The Four Underutilization Rates

• $LU_1$ (Unemployment Rate):

• $LU_2$ (Combined Rate of Time-Related Underemployment and Unemployment):

• $LU_3$ (Combined Rate of Unemployment and Potential Labor Force):

• $LU_4$ (Composite Measure of Labor Underutilization):

$LU_4$ represents the total unmet need for employment in the national economy, capturing the full magnitude of idle and underemployed human potential.

Informal Employment and Decent Work

The ILO defines Informal Employment as jobs lacking basic social protections, written employment contracts, paid annual leave, or severance pay, occurring in either:

• The Informal Sector: Unincorporated household enterprises not registered with national authorities.
• Formal Enterprises: Formal corporations employing informal workers off the books to evade labor taxes.

Official systems track the Decent Work Agenda through indicators including the Gender Wage Gap, the Fatal Occupational Injury Rate, and the NEET Rate (the proportion of youth aged 15–24 Not in Education, Employment, or Training).

Poverty measurement evaluates whether an individual or household possesses sufficient economic resources to attain an acceptable minimum standard of living.

1. Absolute Versus Relative Poverty Lines

• Absolute Poverty Line: Fixed in terms of the purchasing power required to command a defined physical basket of essential goods and services (food, clothing, shelter). The real value of an absolute poverty line remains constant over time, updating only for inflation. Standard in developing countries and for international monitoring (e.g., the World Bank global absolute poverty line of \$2.15 per day at 2017 Purchasing Power Parity).
• Relative Poverty Line: Defined relative to the prevailing economic standard of the society, typically set at $50\%$ or $60\%$ of national median equivalized disposable household income. Standard in the European Union and OECD countries. As society gets richer, the relative poverty line rises automatically.

2. The Cost of Basic Needs (CBN) Method

The Cost of Basic Needs (CBN) approach, widely deployed by NSOs across Asia and Africa, constructs the poverty line in two sequential stages:

Step 1: The Food Poverty Line ($Z_F$)

1. Nutritional experts establish a minimum recommended daily dietary energy intake per capita (typically $2,122\text{ kcal}$ per day, including minimum protein standards).
2. Economists define a normative food basket representing the consumption patterns of the bottom three deciles of the population (rice, pulses, vegetables, oil, fish, milk).
3. The quantities in this food basket are priced using local retail market prices to compute the Food Poverty Line ($Z_F$): the exact monetary cost required to purchase the minimum nutritional basket.

Step 2: The Non-Food Allowance and Overall Poverty Lines

Because even the poorest households must purchase clothing, shelter, and medical care, a non-food allowance must be added to $Z_F$:

• Lower Poverty Line ($Z_{\text{lower}}$): Evaluates the non-food expenditure of households whose total consumption expenditure is approximately equal to the food poverty line ($Y_i \approx Z_F$). Because these households sacrifice necessary food to afford shelter and clothing, their non-food spending ($NF_{\text{minimal}}$) represents an absolute minimal survival allowance:

Households below $Z_{\text{lower}}$ are classified as living in extreme poverty.

• Upper Poverty Line ($Z_{\text{upper}}$): Evaluates the non-food expenditure of households whose food consumption expenditure equals the food poverty line ($F_i \approx Z_F$). Their non-food spending ($NF_{\text{typical}}$) represents a broader social allowance:

Households below $Z_{\text{upper}}$ are classified as living in moderate poverty.

In 1984, Erik Foster, Joel Greer, and Erik Thorbecke formulated an axiomatic family of poverty metrics that unified the entire mathematical theory of poverty analysis.

Let a population consist of $N$ individuals with welfare metrics (income or consumption expenditure) sorted in ascending order:

where $z$ is the official poverty line, and $q$ is the total number of individuals whose income falls strictly below the poverty line ($y_i < z$).

The Foster-Greer-Thorbecke ($P_\alpha$) Metric is defined as:

where $\alpha$ is a poverty aversion parameter governing the ethical sensitivity of the metric to the depth and severity of poverty.

Special Cases of the FGT Metric

1. $\alpha = 0$: The Headcount Ratio ($P_0$ / $H$):

The Headcount Ratio measures the incidence (prevalence) of poverty: the proportion of the national population living below the poverty line. Axiomatic Limitation: $P_0$ violates Sen’s Monotonicity Axiom (if a poor person becomes poorer, $P_0$ does not change) and Transfer Axiom (if income is transferred from a poor person to a richer person, $P_0$ remains unchanged).

2. $\alpha = 1$: The Poverty Gap Index ($P_1$ / $PG$):

where $I = \frac{1}{q} \sum_{i=1}^q \frac{z - y_i}{z}$ is the Income Gap Ratio (the average shortfall among the poor expressed as a fraction of the poverty line). $P_1$ measures the depth of poverty. It satisfies the Monotonicity Axiom. Multiplying $P_1 \times z \times N$ calculates the exact total monetary transfer required to bring every poor citizen precisely up to the poverty line under perfect targeting!

3. $\alpha = 2$: The Squared Poverty Gap (Poverty Severity Index, $P_2$):

By squaring the normalized poverty gap, $P_2$ assigns progressively higher mathematical weight to the poorest of the poor. $P_2$ satisfies both the Monotonicity Axiom and Sen’s Transfer Axiom (a transfer from a poor person to a less-poor person strictly increases $P_2$).

Subgroup Decomposability

A decisive advantage of the FGT family is additive subgroup decomposability. If a population of size $N$ is partitioned into $M$ mutually exclusive demographic or geographic subgroups (e.g., urban vs rural, or 8 administrative divisions) with sizes $N_j$ and poverty metrics $P_{\alpha, j}$:

The percentage contribution of subgroup $j$ to national poverty is $C_j = \frac{N_j P_{\alpha, j}}{N P_\alpha} \times 100\%$, providing policymakers with an indispensable tool for spatial poverty targeting.

While poverty focuses exclusively on the segment of the population falling below an absolute threshold $z$, inequality evaluates the relative dispersion of economic resources across the entire national distribution.

The Lorenz Curve

Max O. Lorenz (1905) formulated the Lorenz Curve $L(p)$, which plots the cumulative proportion of total national income or consumption expenditure received by the bottom $p$ proportion of the population ($p \in [0, 1]$).

Let income $Y$ have cumulative distribution function $F(y)$ with mean $\mu = \mathbb{E}[Y]$. The Lorenz curve is defined as:

Mathematical Properties of the Lorenz Curve

1. $L(0) = 0$ (the bottom $0\%$ owns $0\%$ of income).
2. $L(1) = 1$ (the bottom $100\%$ owns $100\%$ of income).
3. $L(p) \le p$ for all $p \in [0, 1]$: the curve lies entirely below the 45-degree diagonal line of perfect equality ($L_{\text{equality}}(p) = p$).
4. The slope of the Lorenz curve is non-negative and monotonically increasing:

The Lorenz curve is strictly convex to the horizontal axis.

The Gini Coefficient ($G$)

Corrado Gini (1912) defined the Gini Coefficient as the ratio of the area between the 45-degree line of perfect equality and the Lorenz curve (Area $A$) to the total triangular area beneath the equality line (Area $A + B = 0.5$):

Discrete Formulation for Sample Data

For a sample of $n$ individuals with sorted incomes $y_1 \le y_2 \le \dots \le y_n$ and sample mean $\bar{y} = \frac{1}{n} \sum y_i$, the Gini coefficient equals half the relative mean absolute difference:

The Gini coefficient ranges from $0.0$ (perfect equality, every person has identical income) to $1.0$ (perfect inequality, one single person possesses all national wealth while everyone else has zero). In empirical national economies, consumption Gini coefficients typically range from $0.25$ to $0.35$ (low inequality) to $0.45$ to $0.65$ (severe inequality).

Supplementary Inequality Metrics

1. The Palma Ratio: The ratio of the richest $10\%$ of the population's share of gross national income to the poorest $40\%$'s share:

Gabriel Palma observed that middle-income classes ($50\%$ of the population between deciles 5 and 9) consistently capture approximately $50\%$ of national income across all countries; inequality is overwhelmingly driven by the structural tug-of-war between the richest $10\%$ and the bottom $40\%$.

2. Theil’s Entropy Index ($T$): An information-theoretic entropy measure of inequality:

Theil's index decomposes perfectly into within-group inequality and between-group inequality without residual interaction terms.

Human deprivation extends far beyond income or consumption poverty. A household with cash income above the monetary poverty line may still suffer from chronic malnutrition, lack clean drinking water, live in an unpaved mud hut without electricity, and have children out of school.

The Alkire-Foster Multidimensional Poverty Index (MPI)

Sabina Alkire and James Foster (2011) formulated the Multidimensional Poverty Index (MPI), officially adopted by the United Nations Development Programme (UNDP) for global human development monitoring.

The Global MPI evaluates deprivation across three equally weighted core dimensions comprising ten indicators:

1. Health (weight $1/3$): Nutrition ($1/6$), Child Mortality ($1/6$).

2. Education (weight $1/3$): Years of Schooling ($1/6$), Child School Attendance ($1/6$).

3. Standard of Living (weight $1/3$): Cooking Fuel ($1/18$), Sanitation ($1/18$), Drinking Water ($1/18$), Electricity ($1/18$), Housing ($1/18$), Assets ($1/18$).

The Dual-Cutoff Methodology

• Deprivation Cutoff ($z_j$): For each indicator $j$, an individual is defined as deprived ($I_{ij} = 1$) if their achievement falls below $z_j$ (e.g., lacking electricity, or having a malnourished child).
• Deprivation Score ($c_i$): The weighted sum of deprivations suffered by individual $i$:

• Poverty Cutoff ($k$): A person is defined as multidimensionally poor if their deprivation score equals or exceeds a designated threshold $k$ (globally standardized at $k = 1/3 = 33.3\%$):

The Adjusted Headcount Ratio ($M_0$) is the product of two intuitive numbers:

where:

• $H = \frac{q}{N}$: The multidimensional headcount ratio (percentage of the population that is multidimensionally poor).
• $A = \frac{\sum_{i=1}^q c_i(k)}{q}$: The intensity of poverty (the average percentage of weighted indicators in which poor persons are deprived).

$M_0$ satisfies dimensional monotonicity: if a poor person becomes deprived in an additional indicator, $A$ increases, driving up $M_0$.

The UN Sustainable Development Goals (SDG 2030) Monitoring Architecture

The Sustainable Development Goals (2015–2030) comprise 17 Goals, 169 Targets, and 231 unique Statistical Indicators. The global statistical system categorizes SDG indicators into three tiers:

• Tier I: Internationally established methodology exists and data are regularly produced by at least $50\%$ of countries.
• Tier II: Internationally established methodology exists, but data are not regularly produced by countries.
• Tier III: No internationally agreed methodology or standards yet exist (actively developed by custodian agencies).

National Statistical Offices serve as the official national reporting focal points, integrating administrative tax registers, civil registration and vital statistics (CRVS), geospatial satellite imagery, and high-frequency digital mobile surveys into national SDG reporting platforms.

While the standard FGT family provides a foundational benchmark, modern welfare economists deploy sophisticated synthetic metrics that unify poverty incidence, poverty depth, and income inequality among the poor into a single axiomatic framework.

The Sen-Shorrocks-Thon (SST) Index of Poverty

In 1976, Amartya Sen formulated his groundbreaking axiomatic critique of standard poverty measures, demonstrating that neither the headcount ratio ($H$) nor the poverty gap ($PG$) accounts for inequality among the poor themselves. Anthony Shorrocks (1995) and Dominique Thon (1979) refined Sen’s original metric into the Sen-Shorrocks-Thon ($P_{\text{SST}}$) Index:

where:

• $H$: The poverty headcount ratio ($q / N$).
• $I$: The income gap ratio ($\frac{1}{q} \sum_{i=1}^q \frac{z - y_i}{z}$).
• $G_p$: The Gini coefficient of the normalized poverty gaps among the poor population.

Taking logarithms decomposes changes in poverty over time into three distinct policy mechanisms:

This decomposition allows finance ministries to determine whether poverty reductions are being achieved by lifting marginal individuals just below the poverty line out of poverty (lowering $H$), by improving average transfers to the poor (lowering $I$), or by reducing inequality among the destitute (lowering $G_p$).

The Generalized Extended Gini Family ($S\text{-Gini}$)

The standard Gini coefficient assigns linear rank weights $w(p) = 2(1 - p)$ across the population, which implies a fixed, moderate social preference for equality.

S. Yitzhaki (1983) generalized the Gini coefficient into the Single-Parameter Extended Gini ($S\text{-Gini}$):

where $\nu$ is an ethical inequality aversion parameter:

• When $\nu = 2$, $G(2)$ collapses exactly to the standard Corrado Gini coefficient.
• As $\nu \to \infty$, the social welfare weighting shifts entirely to the poorest individual in society, converging to the Rawlsian Maximin criterion ($G(\infty) = 1 - y_{\text{min}}/\mu$).
• When $\nu \to 1$, $G(1) = 0$, reflecting complete social indifference to inequality.

By calculating $G(\nu)$ across multiple values of $\nu \in [1.5, 4.0]$, national statistical agencies provide policymakers with sensitivity bounds that prevent ideological bias in national welfare reporting.