1 Impact of Global Financial Crisis and Assessment of Policy Responses Suzanne Duryea November 18, 2010 presented at UNICEF and CONEVAL Forum on the Impact of the Economic Crisis on Children and Adolescents: Towards and Equitable Recovery This presentation does not represent the official views of the IDB or the board of directors.
Outline I. The Region in Crisis a. Potential Impacts for Children and Youth b. Trends in Regional Indicators c. Policy Responses d. Reflections on Policy Responses II. Impacts of Crisis on Youth in El Salvador a. Motivation/Relevance b. Initial impacts of financial crisis III. Summary 2
a. Potential Effects of Crisis on Children and Youth (based on research from previous crises) Poverty Increases in incidence and depth refs: Lustig 2000; McKenzie 2003 Education Decline in attendance/attainment unlikely (depending on depth) Possible reduction in quality refs: Duryea and Arends-Kuenning,2003; Schady, 2004; Lopez-Boo 2008 Youth Labor Shift in time from school to work unlikely (depending on depth) refs: Duryea and Arends-Kuenning,2003; Schady, 2004 Health* Possible adverse effects on nutrition and infant mortality Declines in health utilization refs: Paxson and Schady, 2004; Baird et. al 2007; Schady and Smitz(2009) * biggest risk for irreversible impacts 3
4 b. Trends in Social Outcomes (relative scarcity of data) Crisis is associated with a reversal of the decline in regional per capita household income poverty in 2009, returning to longer run trend in 2010 50 40 30 40.5 48.3 42.5 43.2 44.0 44.2 42.0 39.8 36.3 34.1 32.1 34.1 30.8 20 10 18.6 22.5 18.1 18.5 19.4 19.1 16.9 15.4 13.3 12.6 12.9 13.7 11.7 0 1980 1990 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 Non-extreme poverty Extreme poverty Source: Forthcoming 2010 Annual Report of the IDB. 2010 data projected from IFM GDP projections and growth elasticity of poverty from 2005 to 2009 with earlier poverty estimates available from CEPAL.
Brazil Argentina Uruguay Panama Peru Chile Guyana El Salvador Costa Rica Venezuela Jamaica Honduras Guatemala Paraguay Ecuador Colombia Mexico Dom Rep Bolivia 5 b. Children are most vulnerable in terms of living in low per capita income households. 35.0% 30.0% 25.0% 20.0% Elderly and youth in the bottom quintile (%) 15.0% 10.0% 5.0% Elderly Youth 0.0% Source: Sociometro, IDB. Household surveys circa 2008 Note: Children defined as <=18 years of age. Elderly defined as >=60 years of age.
6 b. Trends: Crisis not associated with downturns in school attendance for 13-17 year olds (where 2009 data available) Source: CEDLAS.
c. Policy Responses Most prevalent labor market interventions: Deployment of temporary work-fare programs (including Guatemala, Honduras, Mexico, Paraguay, Peru) with much of the work-fare centered in public works/infrastructure. Deployment of Programs to Maintain Employment (Argentina, Chile (for youth 18-24), Mexico (Programa de Preservación del Empleo) Expansions in training programs (Colombia (SENA for youth), Costa Rica, Peru, Mexico) 7
8 c. Policy Responses Most prevalent responses in social protection: Conditional Cash Transfer Programs (CCTs) that increased benefit levels Mexico (Oportunidades) Brazil (Bolsa Familia) Chile (Chile Solidario additional payment to most vulnerable families) CCTs that accelerated expansion Costa Rica (Avancemos) Colombia (Familias en Accion) El Salvador (Red Solidaria) Brazil (Bolsa Familia) Mexico (Oportunidades) Paraguay (Tekoporá ) Jamaica (PATH)
d. Reflections on Policy Responses Improvements over previous crises Most countries entered crisis in good fiscal conditions able to deploy responses more rapidly than in previous crisis avoid precipitous declines in social expenditures seen previously With respect to expanding CCTs many countries were fortunate to have been in the right place at the right time. Launching new programs of this nature in a timely fashion not feasible. 9
d. Reflecting on Policy Responses Region s systematic lack of an automatic micro stabilizer such as unemployment insurance means that countries generally responded in the labor market with more - fragmented policy responses - some of which are more time-consistent - workfare often difficult to dismantle post-crisis - training may be difficult to align with needs of changing economy but can serve well as temporary income support (credible exit strategy) 10
II. Case of El Salvador 11
Why interesting and relevant Exceptional case Should not be taken as representative of region lack of fiscal space proximity to US/depth of crisis Permits an exploration of deep crisis scenario, i.e., extending results in Duryea and Arends 2003 for Brazil We focus on 10-16 year olds because more action between income effect and substitution effect here than in younger age group where substitution effect less applicable 12
El Salvador: Dramatic collapse of growth in 2008 13
El Salvador: Remittance flows also dropped at end of 2008 but returned by end of 2009 14
80 81 82 83 84 85 86 87 Percentage of Children Employed 13 14 15 16 17 18 19 20 15 Unsophisticated analysis/no controls can visually see that crisis disrupted trends in school attendance and employment for 10-16 year olds Mean School Attendance and Employment for 10-16 Year Olds, 2005-2008, July-December 2005 2006 2007 2008 Attendance Employment Source: Author's based on EHPM.
Estimation Strategy: Duryea and Morales We explore school and employment outcomes using linear probability models for a pooled sample of 10-16 year olds over the period of 2000-2008. The repeated cross-sections of data effectively allow us to isolate the effect of the crisis after introducing fixed effects of age, gender, and states with a time trend. For each outcome of Y ist the equation estimated is as follows: 16
Results: note of the first wave of crisis The financial crisis spared no time in taking a toll on children in El Salvador The first wave of the crisis (end of 2008) significantly lowered school attendance of children ages 10-16 by 2.5 percentage points. No differences by gender. The crisis also shifted attendance to the public sector, raising the probability of attending public school by 5 percentage points. No differences by gender With crisis boys are 5 percentage points more likely to be employed, with no effect for girls 17
III. Conclusions The region entered the 2008/2009 crisis in better fiscal shape than previous crises Countries were able to respond aggressively or at a minimum avoid huge contractions in social expenditures seen in previous crises. Numerous social protection instruments rapidly deployed or enhanced. School attendance appears to have been protected in most countries. Other outcomes still pending data availability. An examination of effects of first wave of crisis in El Salvador suggest that youth in countries most profoundly impacted by the crisis may have experienced impacts on education Need for continued research Follow-up on effects of full crisis with 2009 data El Salvador Health impacts 18
19