Cash shortages and consumption during Venezuela’s hyperinflation

ABSTRACT

This paper tests whether cash shortages observed during the hyperinflation in Venezuela reduced consumption. Volumetric consumption indices are constructed for two types of goods: basic and non-basic goods, using data from Nielsen Venezuela at the national level. We find that cash shortages did reduce basic goods consumption but did not affect non-basic goods. The greatest consumption losses occurred between the forced demonetization of December 2016 and the peak of hyperinflation (September 2018). Those losses range between 3 and 10% points (p.p.) of the monthly consumption that would have been observed without the monetary restrictions.

KEYWORDS:

• Cash scarcity
• consumption restriction
• demonetization
• goods shortage
• price control

JEL CLASSIFICATION:

• E21
• E58
• E40

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 from Nielsen Venezuela. Authors could provide the data upon request with the permission of Nielsen Venezuela.

Notes

¹ Out of a total of 44 types of goods, 10 categories were considered basic: edible oils, rice, tuna, corn flour, wheat flour, tomato sauce, milk powder, margarine, mayonnaise, and pasta. In this basket, the consumption of corn flour represents approximately 37% of purchases. Non-basic goods refer to the remaining 34 categories of goods.

² Using this reference or any other does not affect estimations since it only represents a change in the scarcity level.

³ A similar SVAR that only includes monthly inflation and non-basic goods scarcity is estimated for recovering the structural scarcity of non-basic goods. Cash shortage is not included in the system since its coefficients are not statistically significant.

⁴ Christiano (1987) and Sargent (1977) show that under certain conditions in the money market, during hyperinflations, adaptive expectations are equivalent to rational expectations.

⁵ This procedure is equivalent to using a two-stage linear estimation. Current inflation is not employed as an instrument in the first stage because it is highly correlated with consumption. Longer lags of inflation do not add explanatory power for future inflation either.

⁶ A similar linear model to (2) was estimated for non-basic goods consumption. Estimation shows that $\mathit{\lambda }=0$ and stability tests did not reject the null hypothesis of stability. Consequently, we conclude that non-basic consumption does not depend on cash scarcity.

⁷ The identification of structural breaks also deals with the lack of stationarity of some of the variables in (2) by ensuring that estimated residuals are stationary.