The Impact of Lower Oil Prices on Malaysia: A Var Approach1
1. Introduction. This paper examines the implications of lower crude oil prices on Malaysia’s economy. Although Malaysia’s net oil exports are now very small as a share of GDP (0.1 percent in 2013), its gas exports are sizeable (over 6 percent of GDP with gas export prices are linked to crude oil prices through long term contracts. The net effect on the macroeconomy of the decline in oil prices is not clear-cut a priori: Malaysia has an important hydrocarbons exploration, extraction, and processing sector but its economy has diversified, with manufacturing and services now accounting for more than 80 percent of output. While energy exploration and extraction are likely to take a hit from the reduction in oil prices, non-oil sectors (and some oil-related ones) could benefit from lower energy costs and a depreciated exchange rate.
Net Exports of Oil and Gas
2. Source of the shock matters. The reason for the decline in oil prices is also likely to matter. If the recent decline in oil prices is driven more by increased supply of energy rather than by weakening global demand, then Malaysia could potentially be a net beneficiary as the export-oriented sectors of the economy can continue to perform well. Arezki and Blanchard (2014) discuss the respective roles of demand and supply factors behind the recent decline in oil prices concluding that oil market factors, particularly, supply factors are the dominant explanation. In a scenario where a supply shift accounts for 60 percent of the recent decline in oil prices global output is estimated to increase by 0.7 percent in 2015 and 0.8 percent in 2016 compared to the baseline. The effect will be smaller—increases of 0.3 percent and 0.4 percent in 2015 an 2016 respectively—if the supply shift is partly reversed over time.
3. Outline. The paper proceeds as follows. It first provides some background on the structure of energy production and trade in Malaysia. It then presents results from empirical analysis of the oil prices on Malaysia’s growth. This analysis distinguishes between sources of oil price shocks and concludes that the decline in prices is likely to have a net negative impact on growth, even though the recent decline in oil prices partially reflects supply considerations. The final section concludes.
B. Malaysia’s Energy Sector
4. Overview. Malaysia is a high middle income country with a diversified economy dominated by services (55 percent of GDP) and manufacturing (25 percent), with mining (8 percent) and agriculture (7 percent) following in the distance. Malaysia’s economy is highly open to international trade. Exports amount to over 70 percent of GDP, of which three quarters are manufactures and oil and gas are about 14 percent. Domestic energy consumption has until recently been heavily subsidized. This has undoubtedly contributed to the energy intensity of Malaysia’s GDP remaining relatively stable, unlike the declining pattern seen in most advanced economies.
Malaysia: GDP Composition, 2013
Source: CEIC Data Co. Ltd.
Malaysia: Exports Composition, 2013
Total Primary Energy Consumption per Dollar of GDP
Source: US Energy Information Administration
5. Energy Sector. Malaysia’s business and commodity cycles are positively correlated and developments in energy and other commodities are especially important for fiscal and balance of payments developments. The production, processing and exports of energy products (crude oil and products and natural gas) amounted to 15 percent of GDP in 2014 and contributed a net surplus of about 6 percentage points of GDP to the trade balance. State-owned PETRONAS dominates upstream and downstream activity in the energy sector. The next sections provide more details on production and trade in crude oil and natural gas.
6. Background. Malaysia is Southeast Asia’s second largest oil producer after Indonesia. Almost all of its crude oil comes from offshore fields, with most of the reserves located in the Malay basin. Malaysia produces a light and sweet crude, called Tapis. Because Tapis crude oil can produce higher-value products, it is priced higher than other benchmarks. During 2014, Tapis oil averaged about US$10 above the IMF’s average spot price for crude oil. At end-2014, futures markets indicate a Tapis crude average price of US$75 in 2015.
7. Oil production. Malaysia’s crude oil production was about 590,000 barrels per day (bbl/d) during the first 11 months in 2014, down from a peak of about 844,000 bbl/d in early 2000s. The decline was caused by the maturing of oil fields and has prompted Malaysia to seek joint ventures and the government has provided incentives to undertake enhanced oil exploration in order to prolong the life of mature oil fields.
Top Asia-Pacific Proven Oil Reserve Holders, January 2014
Source: Oil & Gas Journal
Malaysia: Petroleum and Other Liquids Production and Consumption, 2000-15
Source: U.S. Energy Information Administration, Short-Term Energy Outlook, October 2014
8. Oil consumption and exports. Rapid economic growth in recent years has raised domestic energy demand and consumption of crude oil. Coupled with declining output, this has resulted in a reduction of crude oil exports. Malaysia remained a net exporter of crude oil and products in 2013, with exports of 240,000 bb/d in 2013 and imports of 183,000 bb/d. It exports sweet crude, which commands a premium, and imports heavier crude oils from the Middle East and other locations for refining and domestic consumption. Primary export destinations are Australia, India, Thailand and Japan.
9. Overview. Natural gas is playing an increasingly important role in Malaysia’s energy production and trade. Its proven gas reserves of 83 trillion cubic feet (Tcf) are the third largest in Asia. Production comes from offshore fields in Sabah and Sarawak and has steadily increased steadily to 2.3 Tcf in 2013. Domestic consumption has also risen to meet the needs of power generation (50 percent) and industry (33 percent). Demand for natural gas for power generation and industrial use is expected to remain strong as Malaysia transitions to high income status. A number of new gas projects are under development. Interestingly, high demand for natural gas in peninsular Malaysia has forced Malaysia to import LNG and invest in regasification.
Malaysia: Dry Natural Gas Production and Consumption, 2000–13
Source: U.S. Energy Information Administration, International Energy Statistics
Top Five Asia-Pacific Proved Natural Gas Reserve Holders, 2014
Source: Oil & Gas Journal, January 2014
10. Natural gas exports. Malaysia is the world’s second largest exporter of LNG after Qatar with exports reaching 1.2 Tcf in 2013. Major destinations are Japan (68 percent), Korea (15 percent), Taiwan Province of China (13 percent) and China (6 percent). Most of Malaysia’s natural gas exports are subject to medium and long term contracts. These contracts stipulate a floor and ceiling price, to protect the seller and the buyer respectively, but are tied to spot crude prices for Japan—the so-called Japan crude cocktail (JCC). In practice, prices for Malaysia’s LNG exports tend to follow those of the JCC with a lag of about 3–5 months. LNG export prices tend to be higher than those for domestic consumers of gas, which are regulated by the government. In addition, PETRONAS maintains a fleet of LNG tankers that meet spot demand for natural gas worldwide.
Malaysia: Crude Oil and 3-Month Lead LNG Export Prices
Source: Bloomberg LP.
C. Empirical Analysis: The Impact of Oil Prices on Growth
11. Overview. The net effect of lower oil and gas prices on growth in Malaysia is not clear-cut a priori. Although there will be a negative impact on the oil and gas sectors, other manufacturing and services will benefit from lower energy costs and a depreciated exchange rate. The positive impact on exports could be further increased by an improved outlook in trading partners, particularly the United States. This section of the paper uses two empirical approaches to examine the potential net impact of lower fuel prices on Malaysia. The first uses a three variable vector autoregression (VAR) model that includes global oil prices, world GDP growth and Malaysian GDP growth. The results from this model suggest a positive (negative) impact on Malaysia from increases (decreases) in oil prices. The first approach does not sufficiently differentiate between alternative types of oil prices shocks and the impact on Malaysia is likely to depend on the underlying cause of the price increases. Therefore, this paper also adopts a second approach which addresses this issue by first identifying different types of oil price shocks using a structural VAR for the oil market; and then assesses the impact of these different types of oil price shocks on the economy. The analysis suggests that even when changes in oil prices are driven by supply considerations, there is a positive correlation with growth in Malaysia, and the net impact of a decline in oil prices is likely to be negative.
12. VAR specification. This VAR includes GDP growth for Malaysia and the United States, and the real oil price. The VAR was estimated using quarterly data, with 4 lags, from 1992Q1 to 2014Q2. The included variables are defined as follows: log change in the real price of oil (measured using the refiner acquisition cost of imported crude oil, from the U.S. Department of Energy, deflated by the U.S. CPI); log change in U.S. seasonally-adjusted real GDP; and log change in Malaysia seasonally-adjusted real GDP. Including U.S. growth into the VAR, in addition, controls for the impact that changes in oil prices have on growth in the U.S., and in addition allows for the analysis of the impact on growth in Malaysia.
13. Empirical results. The impulse responses show that real GDP growth increases in Malaysia with increases in the real oil price (see Figure 1, third row, first column). However, the impact of the increase becomes statistically insignificant after 3 quarters. The analysis also shows that stronger U.S. GDP growth has a positive impact on growth in Malaysia, with the peak effect after about 3 quarters.
Figure 1.Impulse Response Functions from VAR model
Sources: Staff Calculations
Distinguishing Between Sources of Oil Price Shocks
14. Empirical approach. Evidence for the United States shows that the underlying cause of oil prices matters in terms of the impact on economy, see for example, Hamilton (2003), Barsky and Killian (2004), and Blanchard and Gali (2009). These papers distinguish between different sources of oil price shocks and show that that impact on the economy can be different. An implication is that an empirical analysis to assess the impact of oil price shocks should take into account the different types of oil price shocks. Distinguishing between sources of oil shocks is also likely to be relevant for a small open economy, such as Malaysia, that is highly dependent on global trade. In the analysis to follow, the paper makes this distinction using the approach in Killian (2009) which identifies different sources of shocks before assessing their impact on the economy. In the first step, three distinct market shocks are identified; crude oil supply shocks; shocks to the global demand for all industrial commodities; and crude oil market specific shocks. The second step analyzes the impact on growth in Malaysia of each of these different types of oil shocks.
15. Data and sample. In order to distinguish between different types of shocks to oil prices a three variable structural VAR model of the oil markets is estimated using monthly data from 1974:01 to 2014:09. The included variables are: Δprodt, the percentage change in global crude oil production; rpot, the real price of oil; and reat, an index of real economic activity (see charts in Appendix 1). Following Killian (2009), the index of real economic activity provides a measure of the component of worldwide real economic activity that drives demand for industrial commodities and is constructed using dry cargo freight weights.2
16. Oil Market Model. The structural representation of the VAR model for the oil market is:
where εt is a vector of serially and mutually uncorrelated structural innovations. Following Killian (2009),
Shifts in the demand curve for oil are caused either by fluctuations in the global business cycle (aggregate demand shocks) or by changes in the demand for oil that are specific to the oil market, such as weather related shocks or shifts in preferences for holding oil inventories (oil specific demand shock). The structure of
17. Evolution of oil demand and supply shocks. Figure 3 shows the time path of the annual averages of the estimated structural shocks. As is apparent from the figure, at any point in time, the oil market is buffeted by the different types of shocks. For example, a large negative oil supply shock in 1980; a large negative aggregate demand shock in 2008 which occurred along with a negative shock specific to the oil market; and large positive demand shocks—which are unrelated to economic activity—in 1999 and 2000.
Figure 3.Historical Evolution of Structural Shocks
Sources: Staff calculations.
18. Understanding the impact on the Malaysian Economy. The second stage of the analysis addresses the question of what is the impact on the Malaysian economy of the three identified oil price shocks and whether there are differences. The analysis is carried out using quarterly data from 1991:1 to 2014:2. The (monthly) structural innovations from the first stage are averaged for each quarter to compute the quarterly innovations:
In this regression model, the impulse response coefficients at horizon h correspond to ϕjh. The lag length 12 is the maximum horizon for the impulse response function. Two specifications are estimated: one which includes 12 lags of each shock, and a second specification, following Killian (2009), where three separate regressions are estimated, one for each type of shock.
19. Results. Regression results indicate that oil market shocks explain a large portion of fluctuations in Malaysian GDP growth. Figure 4 shows the cumulated response on GDP from each of these shocks, normalized so that each represents an increase in the oil prices. In each chart, the solid line shows the estimated impact from separate models for each shock, as in Killian (2009) while the dotted line shows the estimated impact from a single regression equation including all the shocks. These charts show that there are differences depending on the source of the shock. In the case of an oil supply shock, initially there is a negative impact but over time growth increases. Only in this case, is there a difference between the two specifications, with a much lower impact in the combined model. An increase in oil price caused by higher global economic activity has a sustained positive impact on growth in Malaysia. Finally, increases in the oil demand, unrelated to aggregate economic activity, are also positive for Malaysia but the declines over-time. The chart also includes the impact on the U.S. as a comparator and the results are similar to those in Killian (2009).
Figure 4.Cumulative Impulse Responses to Shocks
Sources: Staff calculations.
20. Conclusions. The results from the empirical analysis suggest that the recent drop in energy prices are expected to have a modest negative effect on Malaysia’s growth prospects in the near term. The analysis presented distinguishes between different sources of shocks to oil prices and assesses their impact on growth in Malaysia. Lower oil prices tend to reduce growth in Malaysia with global aggregate demand shocks having the greatest impact both in the near- and medium-term, as Malaysia benefits from higher oil prices and strong global growth. According to Arezki and Blanchard (2014, the recent decline in oil prices is largely driven by supply factors and despite the potential boost to other sectors from lower energy prices, the net impact is likely to be negative.
ArezkiR. and O.Blanchard2014 “Seven Questions about the Recent Oil Price Slump” iMFdirect Available via the Internet: http://blog-imfdirect.imf.org/2014/12/22/seven-questions-about-the-recent-oil-price-slump/
BarskyR. and L.Killian2004 “Oil and the macroeconomy since the 1970s” Journal of Economic Perspectives18(4): 115–34.
BlanchardO. and J.Gali2009 “The Macroeconomic Effects of Oil PriceShocks: Why are the 2000s so different from the 1970s?” J.Gali and M.Gertler (eds.) International Dimensions of Monetary Policy373–428 (Chicago: University of Chicago Press).
KilianL.2009 “Not All Oil Price Shocks Are Alike: Disentangling Demand and Supply Shocks in the Crude Oil Market” American Economic Review99:31053–1069.
HamiltonJ.2003 “What is an Oil Shock?” Journal of EconometricsElsevier vol. 113(2) (April) 363–398
US Energy Information Administration2014Malaysia ReportSeptember29 Available via at the Internet: http://www.eia.gov/countries/cab.cfm?fips=MY
Prepared by Niamh Sheridan.
The updated series is available on Killian’s website: http://www-personal.umich.edu/~lkilian/paperlinks.html.