Estimation of Elasticity of Intertemporal Substitution. Empirical Monetary Economics

Table of contents

Abstract

List of tables and figures

Table of Abbreviations and Symbols

1. Introduction

2. Theoretical Background of estimating the EIS
2.1 Micro- and Macrodata
2.2 Predictive Power
2.3 Evidence of the Elasticity of Intertemporal Substitution (EIS)

3. Data

4. Empirical results
4.1 Testing the predictive power
4.2 Estimating the EIS
4.3 Tests for Excess Sensitivity

5) Conclusion

References

Tables
Table 1 Baseline Specification from (Crump, et al., 2015, p. 33)
Table 2 Qualitative results of respondent’s predictions
Table 3 Quantitative results of respondent’s predictions
Table 4 Regression of the QTP by demographic-groups
Table 5 Baseline Specification own approach
Table 6 Baseline Specification for different demographic groups
Table 7 Excess Income Sensitivity
Table 8 Excess Stock Price Sensitivity
Table 9 Excess Earnings Sensitivity
Table 10 Long Term Future Inflation Sensitivity
Table 11 Time Dependent Fixed Effects

Appendix
Survey Questions
Conditioning Variables and Specifications
Control Variables: Demos (Categorical)
Control Variables: Test Predictions
Control Variables: Excess Sensitivity
Figure 1 Mean value of QTP4INFL separated by education level & race

List of tables and figures

Table 1 Baseline Specification from (Crump, et al., 2015, p. 33)

Table 2 Qualitative results of respondent’s predictions

Table 3 Quantitative results of respondent’s predictions

Table 4 Regression of the QTP by demo-groups

Table 5 Baseline Specification own approach

Table 6 Baseline Specification for different demographic groups

Table 7 Excess Income Sensitivity

Table 8 Excess Stock Price Sensitivity

Table 9 Excess Earnings Sensitivity

Table 10 Long Term Future Inflation Sensitivity

Table 11 Time Dependent Fixed Effects

Figure 1 Mean value of QTP4INFL separated by education level & race

Table of Abbreviations and Symbols

Abbildung in dieser Leseprobe nicht enthalten

1. Introduction

Especially in times of crisis, such as the current corona pandemic, we are repeatedly reminded that in our economic world all market participants and institutions are deeply interconnected due to dependencies and expectations. Not only central banks, which in recent years have been accused of having a reduced capacity to act as a result of low interest rate policies, but also investors, entrepreneurs and governments are interested in gaining a deeper understanding of the impact of expectations on real value development. From this point of view, deeper questions may arise: “Why are inflation expectations relevant for individual economic decisions?” The literature explains that inflation expectations, as part of real interest rates, can influence investment and consumption decisions (Coibion, et al., 2020, p. 2). The pioneering work of (Hall, 1978), (Hall, 1988) and (Hansen & Singleton, 1982), (Hansen & Singleton, 1983) analyzed this relationship for the Consumption Euler Equation: , (1)
where is the expected consumption growth from t to t+1 ; is the perceived real interest rate from t to t+1 and is the inverse of elasticity term of the intertemporal substitution (EIS). Knowing the latter term, we could potentially classify the effect of perceived real interest rates on expected consumption growth. This implication allows us to see a causal coherence. If the real interest rate increases, current consumption may fall because of the higher return on savings; but current consumption may also rise if the household decides to consume more immediately because they feel richer. This leads us to our research questions: “What is the value of ?” and “What is the effect of perceived real interest rates on expected consumption growth?” Microdata from the New York Fed's Survey of Consumer Expectations (SCE) (Federal Reserve Bank of New York, 2020) will help us to evaluate these questions. Using microdata has the advantage that the process does not have to be based on auxiliary assumptions and that we can estimate the EIS with an ordinary least squares (OLS) estimate rather than using the generalized method of moments (GMM). The heterogeneity of this brilliant database allows us to answer a third interesting research question first: "Does the estimation power of different sociodemographic groups differ?" Integrating the topic of estimation power is a new approach and will also accompany us throughout the empirical examination of the EIS.

In the following chapter we will move from the properties of micro and macro data, individual predictive power, to the estimation results of EIS in the literature. Then we will briefly explain the basic data and discuss our empirical results. We will first test the predictive power, perform the basic regression for estimating the EIS, and then modify it for the excess sensitivity tests. The final section summarizes all results once again.

[...]