Consumption Disasters: A Global Dataset

ABSTRACT

A number of studies suggest that macro-finance models incorporating the risk of extremely large drops in consumption can explain many financial phenomena. This literature is based on the data for a limited number of countries and a relatively small number of disasters in the post-World War II period. Our study constructs a global dataset on consumption disasters. We identify 498 consumption disasters in 212 countries after World War II and provide data on the probability, size, and duration of disasters for the total sample and different groups of countries. We also explore the recovery of consumption after disasters.

KEYWORDS:

• Consumption disaster
• disasters probability
• disasters persistence

JEL:

• E44

Supplemental data

Supplemental data for this article can be accessed online at https://doi.org/10.1080/1540496X.2023.2278661.

Disclosure Statement

No potential conflict of interest was reported by the author(s).

Data Availability Statement

The data that support the findings of this study are available from the corresponding author, upon reasonable request.

Notes

1. A Few studies consider the probability of disaster also as the ratio of number of disasters divided by the total number of observations in the sample (see Barro 2006, 2009). Therefore, we also provide data on this probability.

2. Table A1 in the Appendix provides descriptive statistics for all variables used in the study.

3. The annual frequencies of disasters in more homogenous groups of countries should better represent variations in the disasters risk. However, as consumption disasters are infrequent the inference based on the small group of countries can be questionable (Barro and Ursúa 2012). Therefore, to enable researchers to select the data most suitable for their research Tales A2-A4 also provide data on the number of countries per year in each group. Also, the country level data we provide in the Online appendix enable researchers to construct measures of annual frequency of consumption disasters using different grouping of the countries.

4. Nakamura et al. (2013) use the earlier version of Barro and Ursúa’s (2010) dataset that comprise 24 countries between 1890 and 2006 and 2,685 annual observations.

5. The use of country fixed-effects may cause a dynamic panel bias. However, Teulings and Zubanov (2014) show that the bias declines as T increase and is very small for T = 30 and above.

Additional information

Funding

This work was supported by the Croatian Science Foundation under Grant [IP-2020-02-9710].