Import Tariff and Exchange Rate Transmission in a Small Open Economy

ABSTRACT

We present a small open economy DSGE model with internal and external sticky prices in an incomplete exchange rate pass-through environment. Import tariff is included as another variable that affects the law of one price. The model is calibrated for the Brazilian economy, and the responses of endogenous variables to shocks in import tariff, aggregate supply, monetary policy, and foreign interest are analyzed. The long-run effect of the first shock is deterioration in the terms of trade because the exchange rate appreciation following this shock offsets the initial effect of the increase in import tariff.

KEY WORDS:

• DSGE model
• exchange rate
• import tariff
• small open economy

Notes

^1. Brazil adopted the Mercosur Common External Tariff (CET) in 1995. The Treaty of Asuncion from 1991 required the creation of a common market among Argentina, Brazil, Paraguay and Uruguay. The CET was designed according to a tariff-escalation model, where import duties increase progressively from components or raw materials, through semifinished goods, to finished goods. The CET is on average 12 percent and varies from 0 percent to 20 percent. The decision to raise import tariffs above 20 percent was negotiated within Mercosur and temporarily authorized (MERCOSUL/CMC/DEC. N° 39/11). It was implemented in Brazil according to the CAMEX Resolution n. 70, of September 28, 2012.

^2. The law of one price states that, in the absence of trade barriers, whether natural or imposed by the government, goods should be sold at the same price in all countries when converted to a common currency. However, a large number of empirical studies have found that this law does not hold in practice. The reasons include transportation costs, trade barriers, and anticompetitive market structures (Obstfeld and Rogoff 1996, 202).

^3. Since marginal revenue is negative when the elasticity of demand is less than 1, we require $\epsilon >1$ to ensure an interior solution for the maximization problem of the firm with a positive level of output.

^4. This discount factor came from Equation (11) of the consumer problem, where (1 + i_t) was replaced by the Fisher equation.

^5.

$C_{H,t}^i\left(j\right)=\alpha {\left(\frac{P_{H,t}\left(j\right)}{P_{H,t}}\right)}^{-\epsilon }{\left(\frac{P_{H,t}}{{Є}_{i,t}{P}_{F,t}^i}\right)}^{-\gamma }{\left(\frac{P_{H,t}}{P_t^i}\right)}^{-\eta }{C}_t^i$

^6. The comparison between the simulated impulse-response dynamics and the actual time series of the Brazilian economy is outside the scope of this article. This exercise would require, for instance, the estimation of a vector autoregressive (VAR) model for the observed data, which is beyond the objectives of the article. Whenever possible, we compare our findings with empirical results obtained by other authors.

Additional information

Funding

Author Jose Angelo Divino thanks CNPq from Brazil for financial support.