Small States, Big Effects?

Oil Price Shocks and Economic Growth in Small Island

Developing States

Alrick Campbell

Crawford School of Public Policy, Australian

National University

November 19, 2014

2

Contents

Motivation and Problem Statement

Relation to the Literature

Stylized Facts

Data and Methodology

Main Empirical Results

Conclusions

Motivation and Problem Statement

• Economic viability of small island developing states (SIDS) is under threat– lack of natural resource endowments and high import content (especially of strategic imports such

as fuel)

– inability to influence international prices; and uncertainties of supply due to remoteness and insularity (Briguglio, 1995)

• Considerable scholar focus on oil price-GDP links in developed countries. – Studies mostly find that oil price increases reduce GDP in oil-importing countries

– Effect has become smaller: declining share of energy in GDP, fiscal/monetary policy, exchange rate policy

• Yet, empirical research in this area is limited in SIDS and may contravene theory!– Is it somewhat possible that oil price increases contribute to real GDP growth in SIDS?

– Does high levels of oil and energy intensity imply greater GDP loss?

– What is the role of fiscal policy or the type of exchange rate regime in this regard?

3

Relation to the Literature

• Early Linear Models– Rasche and Tatom (1977) – The 1973-74 oil price shock led to a 7% decline in real GNP in the United States and as

much as 17% in Japan.

• Asymmetry in the oil-GDP relationship– Hamilton (1983); Mork (1989): Rising oil prices seem to retard economic activity more than how falling oil prices

stimulate it.

• Magnitude of the oil-GDP relationship– Blanchard and Gali (2007): A 10% oil price shock would have reduced GDP in the United States by 0.7% prior to 1984

but later declined to 0.25% over a 2-3 year period.

• Studies in SIDS/Developing Countries– Davies and Sugden (2010) - Quantitative analysis of 13 Pacific islands between 2002 and 2009 show that rising oil

prices feed directly into inflation and slow economic growth

– Rasmussen and Roitman (2011) - A 25% increase in oil prices typically causes GDP to fall by about half of 1% or less. Higher oil prices increase the import bill, but effects are partially offset by spending on imports by oil exporters.

4

Stylized Facts – Common Features

Source: United Nations Conference on Trade and Development (UNCTAD)

5

Stylized Facts – Oil Price and GDP Correlation

6

Data and Methodology

Data Set

• Sample – 7 countries– 2 oil exporters – Bahrain and T&T

– 5 oil importers – Cuba, Jamaica, Fiji, Kiribati, Mauritius

– Representative of all regions: Caribbean, Pacific, and Africa,

AIMS

• Annual Data: 1980 - 2012– crude oil prices(lnop)

– exchange rate (lnrer)

– consumer price index (lncpi)

– domestic GDP (lngdp)

– global GDP(lnwgdp)

• Diagnostic checks– All series are I(1): ADF and KPSS Tests

– VAR lag length: BIC

Methodology - Recursive SVAR

– Short-run restrictions on the contemporaneous matrix of coefficients to improve estimation results

𝐵𝑍𝑡 = 𝑖=1𝑝A𝑖𝑍𝑡−𝑝 + 𝑢𝑡,

– 𝑍𝑡 = 𝑛𝑜𝑝𝑖𝑡 , 𝑑𝑙𝑛_𝑟𝑒𝑟𝑡 , 𝑑𝑙𝑛_𝑐𝑝𝑖𝑡 , 𝑑𝑙𝑛_𝑔𝑑𝑝𝑡′

– 𝑛𝑜𝑝𝑖𝑡 = 0, 𝑙𝑛𝑜𝑝𝑡 − ln(max 𝑜𝑝𝑡−1, 𝑜𝑝𝑡−2, 𝑜𝑝𝑡−3 , 𝑜𝑝𝑡−4 )

𝜀𝑡 = 𝐵−1𝑢𝑡

𝜀nopi,𝑡𝜀𝑑𝑙𝑛_𝑟𝑒𝑟,𝑡𝜀𝑑𝑙𝑛_𝑐𝑝𝑖,𝑡𝜀𝑑𝑙𝑛_𝑔𝑑𝑝,𝑡

=

𝛾11 0 0 0𝛾21 𝛾22 0 0𝛾31 𝛾32 𝛾33 0𝛾41 𝛾42 𝛾43 𝛾44

𝑢oil price shock,𝑡𝑢exchnage rate shock,𝑡𝑢inflation shock,𝑡

𝑢domestic demand shock,𝑡

7

Main Empirical Results

• Evidence– Oil price increases generally have a

positive effect on real GDP growth

even after controlling for • Global GDP

• Oil Intensity

• Fiscal Policy

– Most countries utilise a fixed

exchange rate system but evidence

does show that they respond

differently

– Consistent with the findings of

Rasmussen and Roitman (2011)

8

Note: Estimates for individual countries (1980-2012) are from Equation 13 described in Section 4.3.

Bootstrapped standard error bands are based on Monte Carlo simulation (1000 replications).

The solid line is the mean impact and the shaded region represents one standard deviation confidence interval error bands.

Main Empirical Results

• Unambiguous increase in consumer prices

9

• Oil exporters - RER appreciates

• Oil importers - RER appreciates initially followed by a

depreciation by the second year

Conclusions

• Small states and small effects!– After a rise in oil prices, all 7 countries experience a temporary increase in real GDP

– The role of energy intensity and fiscal policy is not relevant in explaining this finding

Possible Explanation:– Exchange rate adjustments: Appreciates for oil exporters and depreciates for oil

importers

– Economic linkages with oil exporters and major economies could be work as natural shock absorber for small open economies: PetroCaribe (LAC)

• Oil price shocks are inflationary in all 7 SIDS economies

• Further Work– Importance of other factors

– Cointegration

10

Thank you for your time and interest.

I look forward to your comments and

questions.

11

Supplementary Material

• GDP to NOPI • Fiscal Policy

12

Supplementary Material

• Oil Intensity • Levels VAR

13