Lars Osberg Economics Department Dalhousie University June 23, 2009 Presentation to: Council of the Federation symposium St. John´s, Newfoundland, June.

Lars OsbergEconomics Department

Dalhousie UniversityJune 23, 2009

Presentation to: Council of the Federation symposium St. John´s, Newfoundland, June 23

Poverty and Recessions: Canada‘s Vulnerable in

Tough Times

2007 – lowest poverty rate in 30 years Statistics Canada no. 75-202-X Income in Canada 2007

Poverty and unemployment - rise and fall with the business cycle

But there has been a long term big change – Poverty is deeper now

So what’s happening to unemployment?

IMF “Advanced economies are suffering their deepest recession since World War II.”OECD Economic Outlook March 2009 http://www.oecd.org/dataoecd/18/1/42443150.pdf

2006 2007 2008 2009 2010

Canada - Unemployment 6.3% 6.0% 6.1% 8.8% 10.5%

So what sort of safety net is there ?

UI/EI Replacement Rate – 38 years of decline The average of the gross unemployment benefit replacement rates for two earnings levels, three family situations Source: OECD, Tax-Benefit Models. http://www.oecd.org/document/3/0,3343,en_2649_34637_39617987_1_1_1_1,00.html ; The Canadian Labour Force Participation Rate Revisited: Cohort and Wealth Effects Take Hold Steven James, Tim Sargent, Russell Barnett and Claude Lavoie Working Paper 2007‐01, Finance Canada, Page 11

EI Replacement Rate: Relatively low by OECD standards

OE C D S ummary Meas ure 2005Averag e Gros s Unemployment B enefit

R eplac ement

2232

41 4935 39

24 33 3526 34

2433

12 1312

S eries 1

The Big Problem The Big Problem – EI is – EI is muchmuch harder to get, especially since harder to get, especially since 19961996

Social Assistance – low & falling ATIONAL COUNCIL OF

WELFARE: WELFARE INCOMES 2005

Alberta Welfare Income: Couple + 2 Kids - 2005 $ - 30% real

cut by provincial SA 1986-2005

Poverty in perspective

How much of market income would it take to completely eliminate the poverty gap ?Canada – 1.7%Nova Scotia – 1.7%

V25746680, v25746752, v25746914; Low income cut-offs after tax, 1992 base; Aggregate low income gap as a percentage of market income (Percent); All family units

For people who do not have much, it does not take much to make a big difference in their lives.

Canada’s total poverty gap - approximately equal to foregone revenue of 2% cut in HST/GST

“Low Income/Poverty Line” - How should we measure it?

Methodologies in use in Canada 1970 – 2008 Note: Because real median incomes flat 1980 - 2006, all effectively updated for inflation only

LICO - % income spent on necessitiesLIM – 50% medianMarket Basket Method

oHRDC or Fraser InstituteSubjective (Leyden)

“Barely adequate for day-to-day living”“Make ends meet”

Appendix G Page 81 Low Income in Canada: 2000-2006 Using the Market Basket Measure HRSDC. October 2008; Table 1, page 12 The Evolution of Poverty Measurement - with special reference to Canada Osberg (2007)

Poverty Line for 2 adult 2 child family (boy 13, girl 9)

$2006

Market Basket Method

LICO IAT rural = 21,860 100-499K = 28,200

500K+ = 30,000

LIM after tax = 29,600

Income Poverty Line in Canada

Broad consensus on approximate level for urban areas for some family types

BUT “Market Basket” Methodology shows importance of Child Care ExpensesTransportation Cost in rural areas – bus service

NOT available

Is Annual Income Inadequacy a good guide for Poverty Policy?

Sen: Commodities are needed for capabilities, which enable valued functioningsPoverty = deprivation of capabilities

Relative income can determine absolute capabilitye.g. When everyone else has a car, how to get

groceries? Capabilities <= $ income + “social wage” + contextProblem: “capabilities” = opportunity set, ≠ observed choicesMeasurement example: % Canadians without car OR bus stop

within 1 KMMultidimensional Poverty indices

Measure achieved functioningsCan look within households

What is critical value of specific item deprivation?How to aggregate over single / multiple deprivations?Correlation of attributes is crucial

Little information added if highly correlated with income, but “too low” correlation implies separable issues involved Measurement error biases towards chance correlation

Social Exclusion – ‘prevented from participation in normal activities of society’

Example of transportation Income poverty – do you have enough cash to buy bus fare ?Capabilities approach

– might ask: is there a bus route? – but capability is fundamentally an individual attribute

Social Exclusion – accessibility planning seen as an issue of community design Bus to where? How often? Is it wheelchair accessible? Where are

services located? Determines feasibility of employment + access to social life &

public services – many feedback effects of isolationMulti-dimensional, mixed indicators, threshold &

feedback effects, long term deprivation crucial Social Exclusion – a relationship of society & the

excludedMeasurement Implication – both personal attributes & social

context of individuals are crucial to social exclusion & poverty

“Everyone has the right to a standard of living adequate for the health and well-being of himself and of his family, including food, clothing, housing and medical care and necessary social services”

UN Universal Declaration of Human Rights (1948) Article

25 Human rights

Specified by Constitutions & International CovenantsClear legal origin, specificity & procedural legitimacy in

democratic institutionsIndivisible & interdependent

E.g. Right to privacy is empty without right to housingImposes obligations on state parties Typically seen as either/or condition

Head-count measure of deprivation ?

Caution:Excess Complexification & its costs !

Main PointsLong term trend increase in Average Poverty Gap in CanadaPoverty Rate moves with Unemployment Rate – rapid rise 2009+Canada’s social safety net cut substantially in mid 1990s – not

fixed

Implications – More Poor People and Deeper Poverty in 2009-11As recession grinds on, faster, deeper income losses likely – but less

social support is now in place for income replacement than in 1990s Work Incentives & Retraining – increasingly irrelevant as jobs

disappearLonger Term Trends

Social Exclusion, Capabilities & Human Rights discourse will widen policy dimensions to include context and “social wage”

Additional slides if time permits

Little change in real median household income in Canada between 1977-2007CANSIM Table 202-0411 –Statistics Canada no. 75-202-X Income in Canada 2007

Was slow growth in living standards inevitable? NOT ! – Canada has worst record

Average Annual Growth of Median Real Income(Equivalised Household Income After Tax)

0.68%

1.50%

0.38%

1.91%

3.62%

2.88%

4.21%

1.15%

3.28%

1.39%

0.15%

0.45%

0.00%

0.50%

1.00%

1.50%

2.00%

2.50%

3.00%

3.50%

4.00%

4.50%

USA 79-04

UK 79-99

SWITZERLAND 82-02

SWEDEN 81-00

SPAIN 80-00

NORWAY 79-00

LUXEMBOURG 85-00

GERMANY 81-00

NETHERLANDS 83-99

FRANCE 81-00

CANADA 81-00

AUSTRALIA 81-03

Average Income has risen because top quintile much better off Statistics Canada 75-202-X Income in Canada 2007

Time and Poverty SpellsAnnual accounting period – too long & too short

No cash & no credit? – very cold in much less than a week

Immediate Needs - historic focus of social policy – now downgraded Human Rights perspective implies short term deprivation matters Stress may trigger events with long term consequences

Long term poverty & inter-generational impacts ? Long term poor – clearly the most deprived – “culture of deprivation” a

real concern Now the focus of “Human Capital” emphasis & “Social Exclusion” discourse

Individuals flow through sequence of householdsPoverty spell entry, exit and recurrence implied by real time

changes in both incomes and household composition Panel data + {assumption: equivalence scale + no transactions costs} can

generate individual life histories of equivalent income poverty spells Costs of volatility & insecurity in health, well-being & human capital not

now recognized

Most Poverty Spells are shortStatistics Canada 75-202-X Income in Canada 2007

But lots of people cycle through Statistics Canada 75-202-X Income in Canada 2007

23-04-19Lars Osberg and Kuan Xu, Dalhousie University, Canada30

Not much change & fairly narrow range – in 2006 $ [Fraser Institute an outlier – “extreme deprivation” concept]

4 person household Low Income/ Poverty Line - urban area of 100-499,000

rural ratio single ratio

Statistics Canada LICO - 1969 base 29,600 1.29 2.2

Statistics Canada LICO - 1978 base 34,400 1.29 2

Statistics Canada LICO - 1992 base (63%) 34,100 1.29 2

Statistics Canada LICO - IAT 1992 base 28,200 1.29 2

CCSD half average income 31,700 1 2.33

Senate Committee (basic needs+30%)*avg growth 35,500 1 2.33

CCSD updated to 2004** 35,200 1 2.33

Senate Committee updated to 2004** 39,400 1 2.33

Statistics Canada subjective -"barely adequate for daily living" 22,700 1 1.4

Statistics Canada subjective -"make ends meet" 35,976 1 1.3

Statistics Canada LIM 50% median market income 29,100 1 2

LIM 50% median before Tax 34,000 1 2

LIM 50% median after tax 29,600 1 2

Sarlo / Fraser Institute market basket 20,320

HRDC market basket* 27,400 0.97 2

Axioms, Aggregation & Dominance – measurement since Sen (1976)

Distribution Sensitive Poverty Indices

SST = FGT1 (1+G(g)) SST = (r) (g) (1+G(gi)).

Average Poverty Gap ratio often ≠ poverty rate

Inequality of poverty gaps is empirically unimportant

Axiomatic basis of Indices Transfer sensitivity axiom

important BUT others ??Focus – relative poverty lines

cannot qualify Impartiality – group identities

of poor irrelevant Continuity – no “threshold

effects” – by assumption Equivalence scales

LIS scale now common Stochastic Dominance of

Deprivation ProfilesRestricted dominance is

relevant criterion Poverty among the Elderly ?

FGTn

z y

z ng

i

qi

ii

q

1 1

11

11

ggg