Mr. Stephan Samuell Statistical Assistant II Central Statistics - - PowerPoint PPT Presentation

• Mr. Stephan Samuell

Statistical Assistant II Central Statistics Office

Day 8: 9 November 2011 Day 8: 9 November 2011

Defining Poverty

Poverty is pronounced deprivation in well‐being, and

comprises many dimensions.

It includes low incomes and the inability to acquire the

basic goods and services necessary for survival with dignity.

Poverty

also encompasses low levels

• f

health and d ti t l t d it ti education, poor access to clean water and sanitation, inadequate physical security, lack of voice, and insufficient capacity and opportunity to better one’s life. ‐ World Bank

Instruments for Poverty Instruments for Poverty Measurement

The Poverty Line

• A poverty line is a monetary measure of the minimum

consumption of goods and services that a household should obtain in order to ensure that its basic needs should obtain in order to ensure that its basic needs are met adequately.

• The poverty line, therefore, represents a minimum

The poverty line, therefore, represents a minimum budget that a household should spend, over a defined time period, if it is to fulfill its basic food and non‐ f d i food requirements.

Instruments for Poverty Measurement

Indigence Line

g

• Minimum Food Basket (MFB): A basket of goods

l d h ’ l selected in such a way as to maximize one’s nutritional needs at the lowest possible cost is referred to as the minimum food basket.

The Indigence Line: also called the food poverty line is

d fi d h f h Mi i F d B k defined as the cost of the Minimum Food Basket.

• Households below the indigence line are considered critically

poor or indigent. The indigence line represents a lower limit p g g p

• f poverty and is the critical poverty line.

Types of Poverty Lines yp y

Absolute Poverty

• Absolute poverty measures the number of people living below a

t i i th h ld th b f h h ld bl t certain income threshold or the number of households unable to afford certain basic goods and services.

Relative Poverty

• Relative poverty measures the extent to which a household's

financial resources falls below an average income threshold for the economy. the economy.

Subjective Poverty

• Basic needs are defined according to people’s perceptions of their

ll b i

• wn well‐being
• The rationale for examining subjective poverty is that people’s
• wn knowledge of their financial situation should be given some

credit.

Poverty Line : Resources

• The poverty line uses economic resources as a welfare

indicator – either income or consumption. C i E di i di

• Consumption

Expenditure is expenditure

• n

consumer goods and services acquired for the satisfaction of wants and needs of households and satisfaction of wants and needs of households and individuals:

1.

Through direct monetary purchases in the market. g y p

2.

Through barter, income in kind

3.

From production within the household (own account production)

Poverty Line : Units of Analysis Poverty Line : Units of Analysis

The main unit of interest : Individuals

k h f ‐ we want to know what percentage of persons are poor

A lot of information is captured at the level of the household A lot of information is captured at the level of the household A household: comprises one or more persons living

p p g together, who occupy the whole or part of a dwelling unit and share at least one of the daily meals.

The households are identified as poor or non‐poor. Thus,

poor individuals , are persons living in a poor household. poo d dua s , a e pe so s g a poo

• use o d.

Poverty Line : Equivalence Scales Poverty Line : Equivalence Scales

The common approach when dealing with

income/expenditure and poverty lines is to express them in income/expenditure and poverty lines is to express them in ‘per capita’ terms

However, this implies that the cost of satisfying each

household member’s needs are the same – this is not true!

Households differ in composition and size‐ thus

consumption by itself is quite misleading with respect to the consumption by itself is quite misleading with respect to the welfare of individuals

To combat this difference in in household composition –

h h ld ‘ li d’ i i l l households are ‘normalized’ using equivalence scales

Such a scale measures the number of adult males the

household is deemed to be equivalent to – essentially q y bringing all households to a ‘leveled playing field’

d l l l Adult Equivalence Scale

Age Male Female < 1 .270 .270 1 to 3 .468 .436 to 3 .468 .436 4 to 6 .606 .547 7 to 9 .697 .614 8 6 10 to 14 .825 .695 15 to 18 .915 .737 19 to 29 1.00 .741 30 to 60 .966 .727 61 + .773 .618

The Cost of Basic Needs Method The Cost of Basic Needs Method

The Poverty Line is comprised of two components, specifically

F d C t ( I di Li ) d N F d a Food Component (or Indigence Line) and a Non Food component.

Basic needs for survival, one would commonly identify :

y y Food, Clothing and Shelter.

• Individuals must eat to survive
• Have some sort of basic clothing and
• Have some sort of basic clothing and
• Occupy some shelter to protect them from the elements

Therefore, in essence, the poverty line is an attempt to

determine the minimum food requirement necessary for healthy living in addition to minimum non‐food healthy living in addition to minimum non food requirements.

Computing the Food Component Computing the Food Component

The Caribbean Food and Nutrition Institute (CFNI) provided the

CSO with a Food Nutrient‐Cost program, called FOODPROG

This

particular program employs mathematical linear This particular program employs mathematical linear programming techniques ‐ designed to calculate the cost of a Minimum Food Basket

FOODPROG contains a database of food items predetermined

by the CFNI ‐ requires the user to input data prices for said food items

Consumer Price Index (CPI) was consulted and a list of items

corresponding to the FOODPROG database was obtained p g

Computing the Food Component Computing the Food Component

Average food prices were calculated for the food items in

fifteen (15) administrative areas geographically spread across Trinidad and Tobago

HBS was carried out during May 2008 – April 2009, we

HBS was carried out during May 2008 April 2009, we essentially had a list of food items containing only 2008/2009 weights and prices. U i h f d i f ll h ll h

Using the average food prices for all these areas as well as the

various area weights, a national weighted average price for each food item was achieved.

These were the final prices entered into the FOODPROG

software which would produce a basket of food items ‐ required by an adult; to achieve a diet of 2400 kcal per day at required by an adult; to achieve a diet of 2400 kcal per day at the most minimum cost.

Computing the Food Component :Minimum Food Basket Composition

Computing the Food Component Computing the Food Component

Daily minimum food basket costs $ 8.22 TT dollars.

Th thl I di Li i $ 8 * $ TT d ll

The monthly Indigence Line is: $ 8.22 * 31 = $ 255 TT dollars Food component = Food Poverty Line : Calculated!

Food component Food Poverty Line : Calculated!

C i h N F d C Computing the Non‐Food Component

Unlike with Food PL, there are no clear parameters to help

answer ‘how much is enough?’ with respect to non‐food items items

Common approach : the non food component is defined as

expenditure of non‐ food items within the framework of f (RG) some reference group (RG)

• Typical reference groups :

1.

Poorest 2 quintiles of data based on HPCAP/Total Expenditure q p

2.

Subset of HBS data set whose total Income/expenditure = FPL

3.

Subset of HBS dataset whose total food expenditure = FPL

Computing the Non‐Food Component

Non‐Food component is derived from manipulation of

• od co

po e t s de

ed

• a

pu at o o the HBS data and can be calculated using several approaches:

The CSO has embarked upon 2 approaches :

1.

Average Non‐Food Expenditure of the poorest 40% of sample

2.

Reciprocal of Food Share Method

Non‐Food Component: Ave. Non‐Food Expenditure Approach

Sort dataset on HPCAP or Total Expenditure Rank data into the various quintiles Rank data into the various quintiles

• 1st quintile being the ‘poorest’ and the 5th quintile representing

the ‘wealthiest’

We require poorest 40% ‐ that is ‐ quintiles 1 & 2 Via SPSS ‘select if’ commands exclude quintiles 3,4 and 5 – so

that we obtain our required poorest 40% reference group that we obtain our required poorest 40% reference group.

We perform further analysis using this subset of the original

data!

Non‐Food Component: Ave. Non‐Food Expenditure Approach

Now the HBS has an original sample 7090 households

• t e

S as a o g a sa p e 7090

• use o ds

Thus each quintile consists of 7090/5 : 1418 households We aggregate the variables corresponding to ‘Total

Non‐Food Expenditure’ and ‘Adult Equivalence Value’ for the 2836 households in the 40% data subset.

To find the average non‐food expenditure: we divide

Total Non Food E penditure/ Total Adult Equi alence Total Non‐Food Expenditure/ Total Adult Equivalence

This value is added to the indigence line to obtain the

poverty line poverty line.

Non‐Food Component: Ave. Non‐Food Expenditure Approach

Summary Calculations Example:

Su a y Ca cu at o s a p e:

Total Food Expenditure= $ 2,824,287.67 Total Non Food Expenditure = $ 6,646,569.80 TOTAL Expenditure = $ 9,470,857.47 Adult Equivalence Sum = 6589.30 q 5 9 3 Non‐Food Component = 6646569.80/6589.30 = $ 1008.70 (monthly) Recall Food Component = $ 255 00 Recall, Food Component = $ 255.00 Then the Poverty Line = 1008.70 + 255 = $ 1263.70

d l f d h Non‐Food Component: Reciprocal of Food Share Method

This methodology for constructing the poverty line –

h b th f d t li i tt i d i th whereby the non‐food poverty line is attained via the inverse of the food share is commonly referred to as the Orshansky Method.

Named after Mollie Orshansky who pioneered this

approach to measure poverty in the USA during the 1960s O h k C ffi i l di / f d

Orshansky Coefficient = total expenditure/ food

expenditure

d l f d h Non‐Food Component: Reciprocal of Food Share Method

Sort dataset on HPCAP or Total Expenditure Rank data into the various quintiles

t

i il b i h ‘ ’ d h

th

i il

• 1st quintile being

the ‘poorest’ and the 5th quintile representing the ‘wealthiest’

We require poorest 40% ‐ that is ‐ quintiles 1 & 2

We require poorest 40% that is quintiles 1 & 2

Via SPSS ‘select if’ commands exclude quintiles 3,4 and 5 –

so that we obtain our required poorest 40% reference group.

We perform further analysis using this subset of the

• riginal data!
• riginal data!

d l f d h Non‐Food Component: Reciprocal of Food Share Method

Find the sum of Food Expenditure and TOTAL

Expenditure for the households within our poorest 2 quintiles quintiles.

Calculate the proportion of food as a proportion of

Total Expenditure – that is – ‘the food share’ Total Expenditure that is the food share

Obtain the reciprocal of the food share value To derive the poverty line we multiply the indigence

To derive the poverty line we multiply the indigence line by the Orshansky Coefficient ‐ Poverty line = Indigence line * Orshansky Coefficient

Non‐Food Component: Reciprocal of Food Share p p Method

Summary Calculations Example: Summary Calculations Example:

Total Food Expenditure= $ 2,824,287.67 p 4 7 7 Total Non Food Expenditure = $ 6,646,569.80 TOTAL Expenditure = $ 9,470,857.47 Food Share = 2824287.67/9470857.80 = 0.30 Reciprocal (Food Share) = Orshansky Coefficient = 1/0.30 = 3.33 Poverty Line = Indigence Line * 3.33 = $ 255 * 3.33 55 3 33 = $ 849.15

P H d R i Poverty Headcount Ratio

The poverty headcount ratio is the proportion of the

i l l i h i b l h ffi i l national population whose incomes are below the official threshold set by the Government – the national poverty line.

Once the number of households which are below the

poverty line have been estimated, the number of people in those households is aggregated to estimate the percentage those households is aggregated to estimate the percentage

• f the population below the line.

Headcount Index is – ideally a measure of prevalence of

y p poverty – it is easy to understand, interpret and communicate.

M t ft d th b i f i ll

Most often used as the basis for assessing overall

progress in reducing poverty.

Computing the GINI Coefficient

The Gini coefficient is a measure of inequality of a

• distribution. In this case we are utilizing the

ffi i f i li coefficient as a measure of income equality.

Where 0 corresponds to perfect income equality –

everyone has the same income everyone has the same income

and 1 corresponds to perfect income inequality – one

person has all the income, and everyone else has zero p , y income.

Computing the GINI Coefficient Computing the GINI Coefficient

Calculation

Ca cu at o

Step 1: Sort the income distribution by income level

(y).

Step 2: Calculate the cumulative distribution function

F(y).

Step 3: Calculate the two essential parameters to

apply the covariance formula: the covariance between income levels and the cumulative distribution function income levels and the cumulative distribution function and the mean income level.

Step 4: Apply Formula: G = Cov(y, F(y)). 2/ (y‐bar)

p 4 pp y (y, (y)) (y ) (where y‐bar: mean of y)