The 2008 Chilean pension reform: household financial decisions and gender differences

ABSTRACT

We evaluate the effect of the 2008 pension reform in Chile, applying a difference-in-difference estimation method to longitudinal survey data representative of the Chilean population. Our evidence suggests that those who started to receive a basic pension increased their debt more than their assets. We interpret this as an indicator of debt sustainability. The debt ratio increased significantly more for women, who may be particularly exposed to financial crises. The results raise concerns about the potential financial vulnerability of the population targeted by the reform, due to over-indebtedness.

KEYWORDS:

• Chile
• pension
• financial vulnerability
• gender vulnerability
• difference-in-difference

JEL CLASSIFICATION:

• I38
• D14

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1. Result of this interest is the creation of the Ministerio de la Mujer y la Equidad de Género of Chile, in 2015, which has among its functions “to promote, coordinate and evaluate the incorporation of the gender perspective in the policies and plans of the various ministries and services at the national and regional levels”.

2. The subsidy was gradually extended, reaching the poorest 60% of the elderly population by the end of 2011, as determined by a poverty score called the “Ficha de Proteccion Social”.

3. Entitlement to the benefit includes at least 20 years residency and residence in the country for at least four of the five years prior to the reform (OECD 2017).

4. Unfortunately, we were not able to use EPS waves after 2009. In particular, the Subsecretaría de Previsión Social describes the 2012 wave as an “incomplete product” and does not recommend its use for statistical inference purposes, due to the level of error this may entail. See www.previsionsocial.gob.cl/sps/biblioteca/encuesta-de-proteccion-social/bases-de-datos-eps/. Data prior to 2006 are reliable, but include information only on a subset of assets, liabilities and income sources in our sample. We use 2004 data in some robustness checks.

5. As robustness checks, in our analysis we also consider two alternative control groups comprising individuals who do not meet the requirements for elegibility, i.e., individuals in the age range 60–65 and individuals who did not fill in the FPS. The control group may not have received PBS because households did not apply (probably unaware of the opportunity) or they were already receiving other pensions from the government. We do not believe any households failed to apply because uninterested in accessing assets and debt markets. The PBS provides a monetary incentive, no matter how it is used.

6. Monetary values reflect 2009 prices adjusted by the consumer price index (source: https://inflationdata.com) and converted into US dollars (source: https://www.investing.com).

7. A problem here is that the questionnaire changed in the two waves, and for year 2004 we have information on a more limited set of income components. Specifically, we do not have information on means-tested family subsidies, water, unemployment and solidarity subsidies that in 2006 accounted for about 20% of median income. The DiD regression thus defines income based on the information available in both waves (earnings for paid work, pensions, food alimonies, donations, helps from relatives and non-relatives).

8. Following the OECD definition, available at (accessed June 12 2019):

https://www.oecd.org/sdd/na/statisticalinsightswhatdoeshouseholddebtsayaboutfinancialresilience.htm.

9. The general aim of the vulnerable or poor filling in the FPS is to gain access to subsidies and benefits from the government (Herrera, Larrañaga, and Telias 2010). We assume that those who did not fill in the FPS are more affluent and anticipate high FPS scores, excluding them from the benefits.

Additional information

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.