The Role of Heterogeneous Expectations in Life Cycle Models

Evaluating the Accuracy of Counterfactuals

Research output: Working paperOther research output

21 Downloads (Pure)

Abstract

This paper shows that subjective survey information on survival expectations can be used to improve the out-of-sample predictions of a dynamic structural model of labor supply, benefit claiming and saving. We consider three approaches to model survival: life tables, average subjective expectations and individual-specific estimates based on reported survival probabilities. The models are estimated on Dutch data from the 1990s, a period during which workers could retire from age 59 at no actuarial penalty to pension benefits. Such actuarial adjustments were introduced in the early 2000s and we use data from the period 2006-2016 to evaluate the accuracy of the counterfactual predictions. While the three models yield different preference estimates, their within- sample fit is similar. Out-of-sample forecasts do differ markedly. Both models based on fixed expectations anticipate a 5-year increase in the average retirement age in the new regime, compared with an observed increase of 2.6 years. The model with heterogenous expectations, on the other hand, predicts a more realistic increase of 2.5 years. We conclude that expectations matter when it comes to counterfactual predictions, even if different combinations of preferences and expectations appear equivalent within a given institutional setting.
Original languageEnglish
Place of PublicationTilburg
Number of pages74
Publication statusSubmitted - 15 Feb 2019

Fingerprint

Life-cycle model
Heterogeneous expectations
Prediction
Out-of-sample forecasting
Survival probability
Survival model
Penalty
Dynamic structural model
Labor supply
Workers
Pensions
Retirement age
Life table
Subjective expectations

Keywords

  • subjective expectations
  • life cycle model

Cite this

@techreport{25d04a47102047ebba94487aea2f9885,
title = "The Role of Heterogeneous Expectations in Life Cycle Models: Evaluating the Accuracy of Counterfactuals",
abstract = "This paper shows that subjective survey information on survival expectations can be used to improve the out-of-sample predictions of a dynamic structural model of labor supply, benefit claiming and saving. We consider three approaches to model survival: life tables, average subjective expectations and individual-specific estimates based on reported survival probabilities. The models are estimated on Dutch data from the 1990s, a period during which workers could retire from age 59 at no actuarial penalty to pension benefits. Such actuarial adjustments were introduced in the early 2000s and we use data from the period 2006-2016 to evaluate the accuracy of the counterfactual predictions. While the three models yield different preference estimates, their within- sample fit is similar. Out-of-sample forecasts do differ markedly. Both models based on fixed expectations anticipate a 5-year increase in the average retirement age in the new regime, compared with an observed increase of 2.6 years. The model with heterogenous expectations, on the other hand, predicts a more realistic increase of 2.5 years. We conclude that expectations matter when it comes to counterfactual predictions, even if different combinations of preferences and expectations appear equivalent within a given institutional setting.",
keywords = "subjective expectations, life cycle model",
author = "{de Bresser}, Jochem",
year = "2019",
month = "2",
day = "15",
language = "English",
type = "WorkingPaper",

}

TY - UNPB

T1 - The Role of Heterogeneous Expectations in Life Cycle Models

T2 - Evaluating the Accuracy of Counterfactuals

AU - de Bresser, Jochem

PY - 2019/2/15

Y1 - 2019/2/15

N2 - This paper shows that subjective survey information on survival expectations can be used to improve the out-of-sample predictions of a dynamic structural model of labor supply, benefit claiming and saving. We consider three approaches to model survival: life tables, average subjective expectations and individual-specific estimates based on reported survival probabilities. The models are estimated on Dutch data from the 1990s, a period during which workers could retire from age 59 at no actuarial penalty to pension benefits. Such actuarial adjustments were introduced in the early 2000s and we use data from the period 2006-2016 to evaluate the accuracy of the counterfactual predictions. While the three models yield different preference estimates, their within- sample fit is similar. Out-of-sample forecasts do differ markedly. Both models based on fixed expectations anticipate a 5-year increase in the average retirement age in the new regime, compared with an observed increase of 2.6 years. The model with heterogenous expectations, on the other hand, predicts a more realistic increase of 2.5 years. We conclude that expectations matter when it comes to counterfactual predictions, even if different combinations of preferences and expectations appear equivalent within a given institutional setting.

AB - This paper shows that subjective survey information on survival expectations can be used to improve the out-of-sample predictions of a dynamic structural model of labor supply, benefit claiming and saving. We consider three approaches to model survival: life tables, average subjective expectations and individual-specific estimates based on reported survival probabilities. The models are estimated on Dutch data from the 1990s, a period during which workers could retire from age 59 at no actuarial penalty to pension benefits. Such actuarial adjustments were introduced in the early 2000s and we use data from the period 2006-2016 to evaluate the accuracy of the counterfactual predictions. While the three models yield different preference estimates, their within- sample fit is similar. Out-of-sample forecasts do differ markedly. Both models based on fixed expectations anticipate a 5-year increase in the average retirement age in the new regime, compared with an observed increase of 2.6 years. The model with heterogenous expectations, on the other hand, predicts a more realistic increase of 2.5 years. We conclude that expectations matter when it comes to counterfactual predictions, even if different combinations of preferences and expectations appear equivalent within a given institutional setting.

KW - subjective expectations

KW - life cycle model

M3 - Working paper

BT - The Role of Heterogeneous Expectations in Life Cycle Models

CY - Tilburg

ER -