HLEG thematic workshop on Economic Insecurity, 4 March 2016, New York, United States. More information at: http://oecd/hleg-workshop-on-economic-insecurity-2016
HLEG thematic workshop on Economic Insecurity, Lars Osberg, presenter
1. ECONOMIC INSECURITYAND THE OPTIMAL
AGGREGATION OF INFORMATION:
COMPOSITE INDICES VS. INTEGRATED
MEASURES
Lars Osberg
High-Level Conference on Economic Insecurity,
OECD/Ford Foundation
New York , March 4, 2016.
2. Economic Insecurity: the anxiety produced by a lack of
economic safety – i.e. by unavoidable downside risk
• How Best to construct an index ?
• [A] Summarize by one variable? OR
• [B] Add Up Insecurities caused by specific future hazards?
• If [B] – which hazards ? How many ? How to aggregate ?
• Economic Insecurity = forward looking concept
• BUT: Past experiences predict current anxieties & probabilities of
future events
• .
3. Optimal Information Aggregation
Why are we doing this anyway?
• Useful for policy or analysis?
• Increase or Decrease in a Single Index is uninformative – gives no hint
as to what has caused change
• Anxieties start with specific worries & large literatures on security of
housing, jobs, employment, food, energy, retirement incomes, etc. exist
• BUT Large & Eclectic “dashboard” can easily drown reader in numbers
• Selection essential but academic whim not necessarily persuasive
• All social measurement aggregates information – what criteria
for a measure of insecurity ?
• A Priori Defensible: Universal; Monotonic in individual hazards
• Cognitively Manageable & Potentially Useful: multiple but not too many
• Process Legitimacy – more than just a random academic’s musings
• Micro Data Based – enable analysis inequality & incidence of insecurities
4. CurrentlyAvailable Measurement Strategies
Single Variable
1.“Economic Security Index”; Jacob Hacker et al
• % Americans experiencing large (>25%) ‘available income’ loss & small wealth buffers
2. Fluctuating Wealth: D’Ambrosio: + Bossert; +Rohde
- wealth & other entitlements protect from future loss, but past wealth losses
increase uncertainty & anxiety about future
3. Income losses relative to personal trend: Rohde/Tang/Rao
- greater past negative shocks relative to trend increase anxiety about future
Compound Index
4. Human Right to Security from specific Hazards: Osberg/Sharpe
- (Now) Aggregates macro estimates of security from hazards of unemployment, financial
costs of illness, poverty for female single parents and poverty in old age
- could be estimated from micro panel data as E (loss │hazard, personal characteristics)
5. “Economic Security Index” : Hacker et al
• “the degree to which individuals are protected against hardship-
causing economic losses.”
• “Large” Current Income loss is measured variable
• % persons experiencing >25% decline in “available household income”
(after medical expenses & taxes) – equivalized for household size
• & without liquid financial wealth > median loss of similar individuals
• Issues:
• “Protection” not measured – social insurance role of welfare state omitted
• No distinction made between market income shocks & transfer responses
• Income declines = choices or constraints ?
• Equivalent income changes with household size changes, whatever cause
• Working Age only – old age security: income uncertainty & care needs?
• Housing wealth not considered – the major asset; leverage + volatile price matter
6. The Buffering Role of Private Wealth:
d’Ambrosio & collaborators
• “the anxiety produced by the potential exposure to adverse events
and by the anticipation of the difficulty in recovering from them.”
• Wealth gives people buffers against future shocks
• α < β - past negative shocks are weighted more heavily than past positive
shocks in calculating the security value of current wealth
• In principle – wealth = entitlements of all types
• In practice – wealth = net private assets, including housing
• House price booms/busts + leverage will dominate middle class net worth fluctuations
• BUT no distinction between choice or constraint in asset change
• PLUS many have nil/negative wealth (Canada 2005 – 24%)
• Hitting lower bound implies zero change over time
• Large spike @ zero in US & Italian data – but their insecurities should be counted
• Nil assets & nil access to credit ? => variation in welfare state (e.g. UI) & labour
market context (Unemp rate) is crucial to economic insecurity
7. Downside Income Volatility Relative to
Trend: Rohde, Tang and Rao
• “a state of stress or anxiety concerning one's financial future”
• Forward looking but inferred from past data
• Panel data enables person specific time series regression
xit = α + βt + eit
• Insecurity <= Large negative shock relative to trend
• ≠ volatility literature – only negative shocks imply insecurity
• Giving someone a lottery ticket does NOT increase their insecurity
• Choice / Constraint not distinguishable as causes of income declines
• Wealth & possible changes in income needs not considered
8. IEWB Index of Economic Security:
Osberg & Sharpe
• “right to security in the event of unemployment, sickness,
disability, widowhood, old age”
• Article 25: UN Universal Declaration of Human Rights
• Disability risk not examined – no reliable data available
• Population weighted macro methodology (could be micro-based)
• probability (unemployment)*(1-Prob(get UI│U)*E(UI Ben│get UI))
• % of disposable household income spent on health care services not
reimbursed by public or private health insurance
• (probability divorce) * (poverty rate single female parent families) *
(poverty gap ratio)
• poverty rate * average poverty gap ratio of 65+
Process Legitimacy: Democratic ratification of Human
Rights Treaties => social choice of which hazards matter
[Bismarck & Stiglitz/Sen/Fitoussi & others provide similar lists]
9. • Core Hypothesis: people feel economically insecure when
they worry about specific hazards which they cannot
avoid
• But not all choices or risks matter equally for policy purposes
• Human Rights covenants affirm “right to security in the
event of unemployment, sickness, disability, widowhood
and old age”,
• Human Rights Law enumerates the primary economic hazards that
democratic societies have agreed are important.
10. Using the ideal micro panel data sets (1)
…. Expectation (t+1) │personal characteristics (t) + history (t-i)
.
• Costs of Unemployment Hazard for Individuals
• Expectation of financial loss = Prob(U)*Prob(get UI│U)*E(Ben│get UI)
• Similar to Boarini /Murtin but family $ to uncovered just transfer risks in family
• IF ADD Expectation (Future Wage Loss + $ value of Psych Costs)
• Will imply heavier weight to Prob (unemp)
• Costs of Sickness Hazard
Unreimbursed Medical Costs as % household disposable income*
• For poor countries, subtract food expenditure (can be 55% median household income)
+ Loss of earnings as % prior household income
• Estimate (expected loss + random) given personal characteristics
ALSO + Cost (Unmet Medical Needs) ?
• Hard to estimate: Highly complex survey evidence required + estimation
11. Using the ideal micro panel data sets (2)
…. Expectation (t+1) │personal characteristics (t) + history (t-i)
• Cost of “Widowhood” Hazard – household composition risk
• Most gendered of hazards: lower female wages + Prob(custody kids)
• “Women and children are one man away from poverty” Canadian social work proverb
• Risk of loss household male earnings <= divorce/separation + death + jail
• Substantial variation in each hazard across & within nations
• Panel data can track household composition changes & income implied
• Expected Value = Prob(2→1 adult family)* Prob(poverty│single parent)
• Cost of Disability Hazard – two types
1. always disabled: no earnings referent, so assume disability benefit as %
average weekly wage
2. transitions to disability: use prior earnings to estimate % loss
• BOTH 1 & 2 imply insecurity for families
• Micro data imply disabled =non-negligible % population (6% Canadians <65)
12. Hazards: = f(individual (unemployment, illness) or
household (widowhood) or life stage (old age))
• Security in Old Age Hazard <= Income & Needs Risks
• Two Social Policy Perspectives Now Prevalent
• Bismark: issue = loss of accustomed lifestyle – too large % decline
• Motivates large “Replacement Rate” literature
= Retirement income / Working Life Income (70% a common goal )
• BUT VERY hard to estimate actual replacement rates over retirement
horizon (≤ 35 years?) using available data sets
• Many assumptions: portfolio decisions + longevity risk + inflation + fin mkts +
• Beveridge: issue =Probability & Depth Poverty in old age
• Much easier to estimate with short panel or cross-section data
• Income needs risks for elderly: greater & different, at life
stage when labour supply response infeasible
• <= elder care costs + medical events + uncertain longevity
13. Certainly out of time by now but some
interesting slides follow……
• And see also:
• Osberg, L. (2015), “How Should One Measure Economic
• Insecurity?”, OECD Statistics Working Papers, 2015/01,
• OECD Publishing.
• http://dx.doi.org/10.1787/5js4t78q9lq7-en
• Or other papers at:
• http://myweb.dal.ca/osberg/
14. “Economic Insecurity”:
Top 5 Google Scholar definitions
• “Economic insecurity describes the risk of economic loss faced by
workers and households as they encounter the unpredictable events of
social life”; (Western et al, 2012).
• “economic insecurity arises from the exposure of individuals,
communities and countries to adverse events, and from their inability to
cope with and recover from the costly consequences of those events”;
(UNDP, 2008).
• “economic insecurity is the anxiety produced by a lack of economic
safety, i.e. by an inability to obtain protection against subjectively
significant potential economic losses”; (Osberg, 1998).
• “Economic insecurity is perhaps best understood as the intersection
between “perceived” and “actual” downside risk”; (Jacobs, 2007)
• “An individual’s perception of the risk of economic misfortune”; (Scheve
and Slaughter, 2004).
15. Insecurity & Origins of Welfare State:
• “The real grievance of the worker is the insecurity of his
existence; he is not sure that he will always have work, he is
not sure that he will always be healthy, and he foresees that he
will one day be old and unfit to work. If he falls into poverty,
even if only through a prolonged illness, he is then completely
helpless, left to his own devices, and society does not currently
recognize any real obligation towards him beyond the usual
help for the poor, even if he has been working all the time ever
so faithfully and diligently. The usual help for the poor, however,
leaves a lot to be desired, especially in large cities, where it is
very much worse than in the country.”
• Chancellor Otto von Bismarck, March 1884,
• Objective: to maintain Kaiser’s rule & prevent social instability
16. Security spending as % of GDP
0.0
10.0
20.0
30.0
40.0
50.0
60.0
Mexico
Korea
Chile
Turkey
Israel
Australia
Iceland
SlovakRepublic
UnitedStates
Canada
Estonia
CzechRepublic
NewZealand
Poland
OECD
Japan
Slovenia
Netherlands
Norway
Luxembourg
Ireland
Hungary
Greece
UnitedKingdom
Portugal
Spain
Germany
Italy
Austria
Finland
Belgium
Sweden
Denmark
France
Figure 1
SOCIAL AND OTHER GOVERNMENT EXPENDITURES AS % GDP, 2009
OLD AGE SURVIVORS DISABILITY HEALTH UNEMPLOYMENT OTHER SOCIAL OTHER GVT EXP
16
17. United Nations’ Universal Declaration of
Human Rights (1948)
• “Everyone, as a member of society, has a right to
social security.”
• Article 22
• “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,
and the right to security in the event of
unemployment, sickness, disability, widowhood, old
age or other lack of livelihood in circumstances
beyond his control.”
• Article 25
17