References

  • Aghion, P., Bacchetta, P., Rancière, R., and Rogoff, K. (2009). Exchange rate volatility and productivity growth: The role of financial development. Journal of Monetary Economics, 56 (4), 494513.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Beetsma, R., Giuliodori, M., and Klaassen, F. (2008). The Effects of Public Spending Shocks on Trade Balances and Budget Deficits in the European Union. Journal of the European Economic Association, 6(2–3), 414423.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Çebi, C. and C¸ulha, A. A. (2014). The effects of government spending shocks on the real exchange rate and trade balance in Turkey. Applied Economics, 46 (26), 315162.

    • Search Google Scholar
    • Export Citation
  • de Castro, F., Fernandez, L., 2013. The effects of Fiscal shocks on the exchange rate in Spain. The Economic and Social Review 44 (2), 151180.

    • Search Google Scholar
    • Export Citation
  • Eichengreen, B. (2007). The real exchange rate and economic growth. Social and Economic Studies, 56 (4), 720.

  • Hartmann Dominik, Christian Jara-Figueroa, Miguel R. Guevara, Alex Simoes and C´esar A. Hidalgo, 2016, “The structural constraints of income inequality in Latin America,” Integration & Trade Journal, No. 40, June, p.7085.

    • Search Google Scholar
    • Export Citation
  • Harvey, D. I., Kellard, N. M., Madsen, J. B., and Wohar, M. E. (2017). Long-Run Commodity Prices, Economic Growth, and Interest Rates: 17th Century to the Present Day. World Development, 89, pp. 5770.

    • Search Google Scholar
    • Export Citation
  • Hausmann, Ricardo, Dani Rodrik and Andres Velasco, 2008, “Growth Diagnostics,” The Washington Consensus Reconsidered, 324355.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Magud, N., and Sosa, S. (2013). When and why worry about real exchange rate appreciation? The missing link between Dutch disease and growth. Journal of International Commerce, Economics and Policy, No. 04(02).

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Oreiro, J. L., Basilio, F. A., and Souza, G. J. (2014). Effects of overvaluation and exchange rate volatility over industrial investment: empirical evidence and economic policy proposals for Brazil. Revista de Economia Política, No. 34(3), pp. 34769.

    • Search Google Scholar
    • Export Citation
  • Spilimbergo, Antonio, Alessandro Prati and Jonathan D. Ostry, 2009, “Structural Reforms and Economic Performance in Advanced and Developing Countries,” IMF Occasional Paper No. 268 (Washington: International Monetary Fund).

    • Crossref
    • Search Google Scholar
    • Export Citation

Annex I. Augmented Dickey-Fuller Test

The test is run by fitting the annual real GDP series (yt) to the following model:

Δyt=α+βyt1+γ2Δyt2+...+γkΔytk+ϵt

The null hypothesis (β = 0), i.e. GDP follows a random walk, is rejected at 1 percent level for all lag specifications except for k=10. The table below presents the test statistics.

article image

Annex II. Commodity Export Price’s Impact on Inflation and Exchange Rate

This annex compares the impact of commodity export price shocks on domestic price level and exchange rate for Uruguay and for other countries in the region, using a SVAR model. Nine commodity exporting countries in Latin America are included in the comparison: Argentina, Bolivia, Brazil, Chile, Colombia, Mexico, Paraguay, Peru, Uruguay.

The empirical model is estimated based on quarterly data from 2000 to 2019.17 The data source of most variables is the World Economic Outlook (WEO). The commodity export price index for each country is compiled using world commodity prices weighted by the country’s commodity export basket. All variables are logged and first differenced to ensure stationarity.

The empirical model takes the following form:

A0Zt=A1Zt1+μt(1)

Zt is the vector of variables of interest, and

Zt=[cxptexrtytpt]T

where cxpt is the commodity export price index of the underling country, exrt is the local currency to USD exchange rate, yt is the real GDP, and pt is the CPI index. Again, all variables are in the logged and first-differenced form.

A0 is a 4x4 lower triangular matrix, with 1 on the main diagonal, and E(μtμt)=Σ. Equation (1) can be rewritten as

Zt=BZt1+Cϵt

where B=A01A1,ϵt=Σ1/2μt, and therefore E(ϵtϵt)=I.AndC=A01Σ1/2.

Since B1* = [b11 0 0 0], the model specification means that cxpt follows an AR(1) process, i.e. the commodity export price index is contemporaneously only affected by a shock to itself, and not by innovations to any other variables. This is a realistic assumption. Though commodity exports is an important sector in the sample countries, considering their sizes, they should be considered as largely price takers of the world commodity market.

The model (A0, Σ, A1) is estimated country by country, with the sample for each country spliting into two sub-samples- one with positive export price shocks, (∆cxpt ≥ 0), and the other with negative price shocks (∆cxpt < 0). Figures 1 and 2 plot the response of domestic prices to a positive and negative commodity export price shock, respectively, for each country.

Figure 1:
Figure 1:

Response of inflation to positive export price shocks

Citation: IMF Working Papers 2021, 002; 10.5089/9781513564708.001.A999

Response of inflation to positive export price shock, SVAR
Figure 2:
Figure 2:

Response of inflation to negative export price shocks

Citation: IMF Working Papers 2021, 002; 10.5089/9781513564708.001.A999

Response of inflation to negative export price shock. SVAR

Theoretically the impact of a positive export price shock on domestic inflation can be ambiguous. On the one hand, the income effect of the export price shock increases aggregate demand, which puts upward pressure on inflation. On the other hand, the local currency tends to appreciate as export earnings increase, which creates downward pressure on inflation as imports become cheaper. Ultimately, which side wins depends on various economy-specific factors, such as the size of imported components in the domestic consumption basket, fiscal discipline of the public sector, and monetary and exchange rate policy choices.

For the majority of countries in the sample, Uruguay included, a positive commodity export price shock appears to lead to higher inflation. Still, as Figure 3 shows, in terms of magnitude and duration, the response of inflation in Uruguay to positive commodity export shocks is among the largest and most persistent compared to others in the sample. The same cannot be said when the export price shock is negative. As Figure 2 indicates, inflation does not appear to respond to negative export price shocks in a consistent pattern in Uruguay, and the magnitude of the response is smaller compared to the case of positive export price shocks.

Figure 3:
Figure 3:

Response of exchange rate to positive export price shocks

Citation: IMF Working Papers 2021, 002; 10.5089/9781513564708.001.A999

Response of LCU/JSD exchange rate to positive export price shock, SVAR

Figures 3 and 4 plot the response of nominal exchange rate to positive and negative commodity export price shocks.

Figure 4:
Figure 4:

Response of exchange rate to negative export price shocks

Citation: IMF Working Papers 2021, 002; 10.5089/9781513564708.001.A999

Response of LCU/JSD exchange rate to negative export price shock, SVAR

Unsurprisingly, positive export price shocks are shown to lead to exchange rate appreciation in all countries in the sample. However, while in most countries the impact appears to have largely dissipated after 8–10 quarters, it seems more persistent for Uruguay. On the other hand, when the export price shock is negative, even though it leads to exchange rate depreciation in the short run, as it does in other countries, the effect turns the opposite direction after about 6 quarters.

Overall, the estimations show that the price responses to commodity export price shocks are asymmetrical in Uruguay. When the export price shock is positive, the responses of inflation and exchange rate are as expected–inflation increases and exchange rate appreciates, leading to real exchange rate appreciation. But when the export price shock is negative, both the responses of inflation and exchange rate appear ambiguous at best. These results are consistent with the observed behavior of inflation and real exchange rate in Uruguay during and after the commodity price boom in the 2000s.

Annex III. Country Groups

The high-growth peer country group:

article image

The trade rival country group:

article image

Annex IV. Structural Indicator Sources

article image
1

I thank Pelin Berkman and the IMF’s Uruguay country team for helpful discussions. The paper also benefited from conversations with colleagues at Banco Central del Uruguay and at various private sector institutions. I take sole responsibility for the views expressed in the paper.

1

See Annex 1 for the test result.

2

The Banco Central del Uruguay aggregates the ISIC classification for exports, which consists of three broad categories: primary activities, manufacturing industries, and electricity, gas, and water. Products are classified as part of manufacturing as long as there is an element of post-primary value added.

3

Also see the Selected Issues Paper for IMF’s 2016 Article IV Consultation for Uruguay.

4

TFP calculated by Penn World Tables is used.

5

This captures both price and quantity effects.

7

See, for example, the commodity price forecasts of the World Economic Outlook for October 2020.

8

One may argue that some of the decline can be explained by weather shocks. Though that explains the export volatility more than the trend.

9

This is partly due to the long production cycle of the industry from tree planting to pulp manufacturing. Thus in any given year, there is relatively few surprises on the supply side, compared to more traditional agricultural commodities.

10

Also see Uruguay’s Selected Issues Paper on exchange rate and competitiveness, for the IMF 2018 Article IV Consultation.

12

Defined as GDP per capita relative to the US falling within a 10 percent band around the level of Uruguay 2018.

13

Defined as the relative GDP per capita growth being higher than 70 percent of the whole sample.

14

See Annex III for the country/year combinations included in the high-growth peer group and the trade rival group.

15

For cross country comparability, PPP adjusted data from the Penn World Table are used.

16

See the Selected Issues Paper of IMF’s 2021 Article IV Consultation with Uruguay.

17

Estimations were also run using longer time series starting 1980. Most of the results are not sensitive to changes in time coverage.

Dissecting Economic Growth in Uruguay
Author: Ms. Natasha X Che