Since June 2014, oil prices have dropped about 65% in US dollar terms (about $70) as growth has progressively slowed across a broad range of countries. This outcome has puzzled many observers including us at the Fund, who had believed that oil price declines would be a net plus for the world economy (IMF 2015), obviously hurting exporters but delivering more-than-offsetting gains to importers.
The key assumption behind that belief is the assumed difference in saving behaviour of oil importers versus oil exporters.
- If oil importers save 10 cents on the dollar while exporters save 20 cents, transferring a one dollar of income from exporters to importers raise global consumption.
For example, when an oil price fall transfers $1 of income, global consumption rises by 90 cents (the extra consumption of importers) less 80 cents (the consumption reduction by exporters). A net increase of 10 cents in global consumption is the result.
- The conventional belief that an oil price fall is good for the world economy then follows from the claim that for the most part, consumers in oil-importing regions such as Europe have a higher marginal propensity to consume that those in exporters such as Saudi Arabia, which can run down its wealth or borrow to cushion temporary declines in export earnings.
Stock markets are not convinced: Is the low interest rate environment to blame?
Clearly, world equity markets have not bought this theory (Blanchard and Acalin 2016). Over the past six months or more, equity markets have tended to fall when oil prices fall – not what we would expect if lower oil prices help the world economy on balance. Indeed, since August 2015 the simple correlation between equity and oil prices has not only been positive (Figure 1), it has doubled in comparison to an earlier period starting in August 2014 (though not to an unprecedented level).
Figure 1 Stock market and oil prices
(60 days rolling windows of daily return correlation)1
Note: 1 Correlation between WTI oil price and MSCI USA return. Horizontal red lines indicate the average correlation for the corresponding periods.
Past episodes of sharp changes in oil prices have tended to have visible countercyclical effects – for example, slower world growth after big increases (Hamilton 2009). Is this time different? Several factors affect the relation between oil prices and growth, but we will argue that a big difference from previous episodes is that many advanced economies have nominal interest rates at or near zero.
Supply versus demand
One obvious problem in predicting the effects of oil price movements is that a fall in the world price can result either from an increase in global supply or a decrease in global demand. But in the second case, we would expect to see exactly the same pattern as in recent quarters – falling prices accompanied by slowing global growth, with lower oil prices cushioning, but likely not reversing, the growth slowdown.
Figure 2 Commidity price indices
Source: IMF, Primary Commodity Price system.
Slowing demand is no doubt part of the story, but the evidence suggests that increased supply is at least as important. One telling indicator is the different behaviour of the prices of oil and other commodities.
- If oil prices reflect a generalised fall in global demand, then with similarly low price elasticities of supply, we would expect other commodities to show a price path similar to oil’s.
But that is not even approximately the case.
Metals began their decline in 2011 with the start of China’s rebalancing and growth slowdown – China consumes upward of 50% of global supply of some metals – and the decline has been more or less steady (Figure 2). In sharp contrast, the fall in oil began only in the second half of 2014, and it has been much steeper than the decline in other commodities since then.
More generally, oil supply has been strong owing to record high output from OPEC members including, now, exports from Iran, as well as from some non-OPEC countries. In addition, the US supply of shale oil initially proved surprisingly resilient in the face of lower prices. Figure 3 shows how OPEC output has recently continued to grow as prices have fallen, unlike in some previous cycles.
FIgure 3 World oil production growth
(Change in million barrels a day produced, unless noted otherwise)
Note: APSP (average petroleum spot price): simple avg. of UK Brent, Dubai Fateh and West Texas Intermediate.
Source: IMF, Primary Commodity Price Sytem; IEA and IMF staff calculations
Moreover, even in the US – a net oil importer where demand has been fairly strong – cheap oil seems not to have given a substantial fillip to the growth rate. Econometric and other studies (Arezki and Blanchard 2015) suggest that only part of the recent decline in oil is due to slowing demand – somewhere between a half (Baumeister and Hamilton 2015) and a third (World Bank 2015) – with the balance accounted for by increasing supply.
So there remains a puzzle.
Where in the world can the positive effects of lower oil prices be seen?
To address this question, we compare the 2015 and 2016 domestic demand growth in oil importers and oil exporters that we projected in April 2015 – after the first substantial decline in oil prices – to the 2015 outturn and the 2016 projection from the latest, April 2016 World Economic Outlook. The April 2015 forecast for oil prices was slightly above actual prices for that year, but was much higher for 2016 and beyond. As a result, absent other shocks, one would have expected the latest WEO forecasts to show stronger performance in oil importers, especially for 2016, and weaker performance in oil exporters.
- The lion’s share of the downward revision for global demand comes from oil exporters – despite their relatively small share of global GDP (about 12%).
- But domestic demand in oil importers was also no better than we had forecast, despite a fall in oil prices that was much bigger than anticipated.
Understanding why the naked eye cannot detect positive spending effects requires a closer look at the composition of demand in oil exporters and importers.
Domestic demand in oil exporters
In 2015, domestic demand in oil exporters was indeed much weaker than we had forecast a year before, and this year is projected to be no different (Table 1) – indeed, for both years domestic demand is contracting.1 Since part of the decline in domestic demand is reflected in declining imports, especially of capital goods, GDP growth has fared a bit better – it is estimated to have been barely positive in 2015 and is projected to pick up slightly in 2016.
Table 1 GDP growth and growth in domestic demand, April 2016 vs April 2015 WEO
The negative demand surprise reflected both weaker consumption and especially weaker investment.
- Rich oil exporters can draw on their reserves or sovereign wealth funds (Arezki et al. 2015), and most of them have, but they have also been cutting government spending sharply.
- Poorer countries, of course, have much lower borrowing capacity, and risk crises if foreign debt levels get too high.
Most have sharply lower current account surpluses or higher deficits, and their sovereign spreads have risen (see Table 2 and Figure 4). Especially in these countries, domestic spending can fall sharply, in a nonlinear fashion (Krugman 2016).
- Public investment has fallen particularly fast – most capital goods are imported, and when fiscal adjustment is needed, capital spending is typically the first item to be cut.
And of course, factors unrelated to oil prices have also being weighing on economic activity in a number of oil exporters – ranging from domestic strife in Iraq, Libya, and Yemen to sanctions in Russia. Default crises or a deeper economic slowdown could generate negative regional spillovers through remittance, trade, and financial flows. In addition to the other factors we have discussed, the fear of such risks is a negative influence on financial markets and global growth.
Table 2 Current account balances of major oil exporters
Source: World Economic Outlook database, IMF and national sources.
Figure 4 Oil exporters' global bond spreads
(Average JP Morgan EMBIG Sovereign Spread Index)
Note: Oil exporters are comprised of Angola, Bolivia, Colombia, Ecuador, Gabon, Iraw, Kazakhstan, Nigeria, Russia, Trinidad and Tobago and Venezuela.
Source: Bloomberg, LP and IMF staff calculations.
Of course, low oil prices make exploration and extraction activities less profitable for the private sector, leading to lower capital expenditures there as well. According to Rystad Energy, the fall in global capital expenditure in the oil and gas sectors amounted to about $215 billion between 2014 and 2015 – about 1.2% of global fixed capital formation (or just below 0.3% of global GDP). This decline has been a major factor for oil exporters (see Figure 5), but oil-related investment in some oil importers has also been hit hard, notably in the US, which accounts for a significant part of the global drop in energy-related investment.
Domestic demand in oil importers
In principle, the decline in oil prices should have provided a boost to domestic demand in oil importers. But we see generally different outcomes for advanced countries and others.
- The situation for oil importers in the emerging and developing world is varied.
These countries typically have more limited pass-through from international to domestic fuel prices compared with advanced economies. For example, China has experienced a drop in domestic (wholesale) gasoline prices of only about 30% since June 2014, far below the 70% fall in the world gasoline price. True, governments’ improved fiscal positions should eventually result in lower taxes or increased public spending, but the process could take time and is subject to various frictions and leakages.
Still, data and projections from the April 2016 WEO suggest that domestic demand in emerging and developing economies is stronger than forecast a year ago, once Brazil is excluded. And this relative strength is evident despite difficult macroeconomic conditions in a few countries that are exporters of non-oil commodities. However, weaker-than-expected external demand (reflecting both weak demand in oil exporters and lower-than-expected demand growth in advanced economies, as discussed further below) implied slightly weaker GDP growth than the previous year’s forecasts.
- The situation is different for most advanced economies.
For 2015, domestic demand growth was slightly weaker than expected a year ago, and the difference is projected to widen in 2016. And with weaker external demand (including, importantly, from oil exporters), GDP growth is close to a half percentage point weaker in both 2015 and 2016.
As shown in Table 3, domestic demand has strengthened above expectations in the Eurozone. But elsewhere things have been different – domestic demand has been weaker than expected not just in commodity-exporting advanced economies (Australia, Canada, Norway, and New Zealand) but also in advanced Asia as well as the US. While clearly other shocks have been at play – for instance, advanced Asian economies (Hong Kong, Japan, Korea, Singapore, and Taiwan) experienced a significant shortfall in their exports to China relative to our forecasts – understanding these differences also requires taking a look at the breakdown of domestic demand between consumption and investment.
Table 3 Advanced economies: GDP and domestic demand growth, April 2016 WEO vs April 2015 WEO
- Advanced oil importing economies have indeed seen some positive effects on private consumption – albeit marginally less than anticipated a year ago.
In the US, retail gasoline prices (including taxes) dropped from $3.8 to $1.9 per gallon between June 2014 and February 2016, according to the US Energy Information Administration. The Council of Economic Advisers (CEA 2016) estimates that in 2015, the US spent about $100 billion less on imported oil than it would have if prices had stayed at their mid-2014 levels. Assuming all the savings on imported oil were spent, that would add 0.5 (0.8) percentage points to GDP (private consumption) growth in 2015. But how much did consumers really spend out of each dollar of savings from lower oil prices? Published empirical estimates from alternative sources range from 40 cents to 80 cents (Farrell and Greig 2015). A cursory look at actual data confirms that household saving rates have increased in many advanced economies; and while an increase in saving may be associated with household deleveraging that could lead to more robust consumption later, it mutes the potential increase in demand today.
But the main reason for the demand shortfall in advanced economies relative to earlier forecasts has been weaker investment (Figure 5). The decline in investment in energy and mining (and related industries) has played an important role, with a notable downward revision to investment growth in commodity-exporting advanced economies (red line) but also the unexpectedly large decline in US energy-related investment noted above (green line). And, at the same time, there has been no discernible boost to investment outside the oil sector, contrary to what was expected. One piece of the explanation may be that the greater uncertainty accompanying the fall in oil prices raises the option value of waiting (Dixit and Pindyck 1994), even in sectors that benefit from the persistently low oil price. Therefore, investment is postponed.
Figure 5 Real fixed investment growth, advanced economies, April 2016 WEO vs April 2015 WEO
The current low oil (and other fossil-fuel) price environment is also discouraging investment in the renewable sector (Arezki and Obstfeld 2015).
There is another factor potentially impeding a pickup of demand in oil importers.
Surprising macroeconomics at the zero interest-rate bound
Compared with previous price cycles, falling oil prices this time coincide with a period of slow economic growth – so slow that the major central banks have little or no capacity to lower their monetary policy interest rates further to support growth and combat deflationary pressures.
Why does this matter? A large economics literature in the 1970s and 1980s – summarised by Michael Bruno and Jeffrey Sachs (Bruno and Sachs 1985) – showed how oil-supply-driven price increases lead to stagflation, i.e. a combination of higher inflation and slower growth.
- Stagflation is a direct result of higher costs for producers who use energy, costs that lead them to reduce output, shed labour, and raise prices to cover higher costs.
- The reasoning should work in reverse when oil prices fall; cheap oil should lead to lower production costs, more hiring, and reduced inflation – even though oil is a less important production input than it was three decades ago.
But this channel causes a problem when central banks cannot lower interest rates in the face of lower inflation (as they would tend to do in normal times).
Falling inflation raises real rates at the ZLB
Because the policy interest rate cannot fall further once it is at zero, the decline in inflation owing to lower production costs raises the real rate of interest. This compresses demand and very possibly stifles increases in output and employment.
Something like this may be going on at the present time in some economies. Figure 6 is suggestive of a depressing effect of low expected oil prices on expected inflation. It shows the strong recent direct relationship between US oil futures prices and a market-based measure of long-term inflation expectations.
Figure 6 WTI oil and the Fed five-year forward breakeven inflation rate
Note: Updated from Citi Market Commentary, 28 January 2016.
Source: Bloomberg, LP.
Oil price increases may lower real rates
Being near the zero bound also can imply a ‘perverse’ response to higher oil prices.
- When central banks are battling deflation pressures, they are unlikely to raise policy interest rates aggressively to counter an uptick in inflation; thus, oil price increases, symmetrically, can be expansionary by lowering the real interest rate.
Of course, it would be wrong to conclude that central banks can enhance the benefits of current low oil prices by raising their policy interest rates. On the contrary, all else equal, that action would harm growth by raising real interest rates.
- Our claim is simply that when an oil importer’s macroeconomic conditions warrant a very low central bank interest rate, a fall in oil prices could move the real interest rate in a way that runs counter to the positive income effect.
Not all supply shocks are created equal
Finally, going back to the very nature of the price shock, the general category of ‘supply shocks’ masks different economic changes with different potential effects on global growth.
One type of supply shock is a discovery – whether of new oil resources or of a new technology such as hydraulic fracturing that unlocks new potential – that effectively expands the world’s usable stock of petroleum. In principle, such a discovery enlarges global production possibilities and should eventually be expansionary at the global level, notwithstanding uneven distribution effects across countries.
However, some of the recent turmoil in oil markets comes from a breakdown of the OPEC cartel’s ability to support the world oil price. In the resulting price war, countries pump more oil today, pushing the price down, but at the cost of lower supplies, and hence higher price, tomorrow. Supply is indeed higher in the short run, but the price war does not increase the world’s energy resources, and so will necessarily be less expansionary than an increase in usable world petroleum reserves.
The way forward
Persistently low oil prices complicate the conduct of monetary policy, risking further inroads by unanchored inflation expectations. What is more, the current episode of historically low oil prices could ignite a variety of dislocations including corporate and sovereign defaults, dislocations that can feed back into already jittery financial markets. The possibility of such negative feedback loops makes demand support by the global community – along with a range of country-specific structural and financial-sector reforms – all the more urgent.
Arezki, R. and O. Blanchard (2015), “The 2014 oil price slump: Seven key questions”, VoxEU.org, 13 January.
Arezki, R. and M. Obstfeld (2015), “The price of oil and the price of carbon”, VoxEU.org, 3 December.
Arezki, R., A. Mazarei, and A. Prasad (2015), “Sovereign wealth funds in the new era of oil”, VoxEU.org, 29 November.
Baumeister C. and J.D. Hamilton (2015), “Structural Interpretation of vector autoregressions with incomplete identification: Revisiting the role of oil supply and demand shocks”, Working Paper, November.
Blanchard, O. and J. Acalin (2016). “Lower Oil Prices are Good for the United States” in O. Blanchard and A. Posen (eds.), Reality Check for the Global Economy, Peterson Institute for International Economics, Briefing16-3.
Bruno, M. and J.D. Sachs (1985), Economics of Worldwide Stagflation, Harvard University Press, February.
Council of Economic Advisers (2016), Economic Report of the President, Washington, DC, February.
Dixit, A.K. and R.S. Pindyck (1994), Investment under Uncertainty, Princeton University Press.
Farrell, D. and F. Greig (2015), “How falling gas prices fuel the consumer: Evidence from 25 million people”, JPMorgan Chase Institute, October.
Hamilton, J.D. (2009), “Causes and consequences of the oil shock of 2007–08”, Brookings Papers on Economic Activity, Spring.
International Monetary Fund (2015), Global Implications of Lower Oil Prices, IMF Staff Discussion Note 15/15, Washington, DC, July.
Krugman, P. (2016), “Oil goes nonlinear”, New York Times Blog, 16 January.
World Bank (2015), "The Great Plunge in Oil Prices: Causes, Consequences, and Policy Responses", Policy Research Note PRN/15/01, March.
 The table shows growth rates for country groups aggregated using market exchange rate weights. Figures using PPP weights show a very similar picture.