On our estimates, global oil demand will end up increasing by about 3.0mn b/d between 2014 and 2016, well above the recent historical averages. Interestingly, both developed (DM) and emerging markets (EM) have contributed to the rebound in oil consumption, with DMs making up about 20% of the incremental barrels burnt. In fact, demand across OECD economies has experienced the largest bounce since its post financial crisis recovery in 2010, expanding by 400 thousand b/d in 2015 and 200 thousand b/d in 2016. In our view, much of this bounce in OECD consumption can be traced back to a price effect impacting the number of miles driven and the efficiency (or lack thereof) in the use of oil.

While OECD oil demand has experienced a strong run, we see the tide turning soon and project a 120 thousand b/d contraction for next year. Put differently, we believe fuel demand growth in OECD Europe and Asia will soon fade out and return to its structural demand decline. Looking back, it is worth remembering that North America has benefitted from a stronger economy relative to Europe and Japan, and that US oil demand has expanded by an average of 270 thousand b/d annually since 2013. But even in the United States we now see consumption flat in 2017 as price effects start to fade out.

Globally, we see oil demand slowing down to 1.2 million b/d in 2017 and project most of the slowdown to come from OECD economies, while non-OECD oil demand growth should accelerate modestly from 1.2 in 2016 to 1.3 million b/d next year. In part, the slowdown is the result of our projected $61/bbl average Brent crude oil price for next year, as consumers in Turkey, Japan, Germany or the US are relatively price-sensitive. When modeling these effects into an impulse/response function, we find that US oil demand sensitivity to prices is more lagged than oil demand in other OECD countries.

In fact, there is little doubt that consumers in the US and elsewhere have opted to buy bigger cars encouraged by the lower fuel prices at the pump. As a result, the average car in the US is larger today than it was five years ago. True, fuel efficiency standards were pretty much flat until about 10 years ago when they started rising sharply on mandated fuel efficiency improvements. But despite the introduction of tighter fuel efficiency standards, average vehicle efficiency has stagnated in the last few quarters. However, as growth in SUV sales start to stabilize as a share of the fleet, we see fuel efficiencies kicking in again. After all, SUV sales cannot keep on growing forever as a proportion of the overall fleet.

Slower demand across the OECD is a problem for the oil market because crude oil and petroleum product inventories across the OECD remain exceptionally high. Should demand contract at a faster rate than we currently anticipate in 2017, the risk of a storage containment problem could rise once again. It is worth noting that total oil stocks in the United States are 117 million barrels above last year and 256 million above the 5-year average, mostly on the back of both high crude oil and petroleum product inventories. Worryingly, the US has kept on building inventories in recent months even as other regions stabilized.

Total stocks in OECD Europe are also bloated, mostly as a result of exceptionally high gasoline and diesel inventories. Yet European crude oil stocks have started to normalize already, positively impacting Brent crude oil spreads. Lastly, the inventory picture in Asia is also quite bleak, with total stocks in excess of 440 mn barrels, 24 MN above the seasonal normal. Having said that the Asian inventory picture is perhaps somewhat better balanced that the one observed in the Atlantic Basin, helped by robust regional demand.

The necessary clean-up in petroleum product markets will likely take some time but it has started already, in our view. Refinery runs across the OECD have been lower than they were last year on weaker product margins. As we discussed in a recent piece (see The residual oil paradox), 3-2-1 cracks have come down from an average level of $25/bbl in June/July of last year to $12/bbl this year, discouraging runs. Yet refinery maintenance this fall is likely to be relatively light compared to the previous two years of particularly heavy seasons. As such, the oil market will keep relying on a stronger demand picture. But will that continue?

Americans love to drive and should clock in north of 3bn miles on highways this year again on their vehicles. In fact, miles driven have increased by 3.4% YoY yearto-date after surging by 3.5% in 2015, well above the 20 year average of 1.3%. Yet, after increasing at a brisk pace, job creation is likely set to slow as the US approaches full employment. Also, gasoline prices still remain today below seasonal averages, encouraging even more driving. But prices effects could fade if crude oil rallies. We expect US premium gasoline retail prices to average about $3/gal next year, a 15% increase YoY. With GDP set to expand by 2.1% next year, miles driven may just increase by 0.8% YoY, significantly below trend, and not enough to offset efficiency gains, which means US gasoline demand could start contracting again.

Cyclical demand for trucking has deteriorated tremendously in recent quarters, with US truck miles contracting on weak industrial activity and low retail restocking. Yet we see a modest turnaround ahead. After all, oil demand in the US and other economies is closely linked to PMIs and these are firmly above 50 again, with our economists projecting a sequential l improvement in 2H2016 and 2017. In our view, an improving outlook should lend some support to cyclical demand and hence oil prices, likely offsetting softer light duty vehicle sales.

Demand for petroleum products across most major countries Europe has been firm this year despite continued fuel efficiency gains, with jet/kerosene consumption hitting a new record on increased air traffic and demand for diesel expanding too. Overall European demand is set to grow 80 thousand b/d this year, down from growth of 210 thousand b/d last year. Next year we expect a resumption of the declines we've seen in the last decade, with demand contracting 80 k b/d in 2017.

Demand across OECD Asia has been rather muted during the past few quarters with Korea offsetting Japan. Just like in Europe, jet/kerosene and distillate demand continues to lead the way, coupled with NGLs. However, the OECD Asian region is somewhat different as demand has been contracting since the 1990s mostly as a result of Japan's multi-decade economic stagnation. In contrast, Australia and South Korea have experienced a more positive oil demand backdrop albeit from a much smaller demand base.

Looking forward, we expect OECD Asia demand to contract by 50 thousand b/d next year, once again driven by Japan. In part, we expect a meaningful impact on oil consumption following the return of Japanese nuclear power generation capacity over the next three years, and see oil consumed in power generation trending lower over the coming months. On top of this, fuel efficiency gains in the region will likely keep pressing demand lower.

In short, OECD demand will likely remain a drag on global oil consumption next year on weak economic growth, stagnating demographics, higher oil prices, and a reacceleration in fuel efficiencies. In contrast, we see EM oil demand growing by 1.3 million b/d in 2017 even as global crude oil prices start to rise modestly towards $60/bbl. Some investors remain skeptical that EM growth will accelerate next year. Yet our demand growth projections sit right around the 20-year average and simply imply a sequential improvement in South America and the CIS. True, oil prices keep a strong correlation to the USD, and a hawkish Fed is a key risk to our outlook.

Yet interest rates in local currency across many EMs have continued to trend lower for now. The same is true for interest rates on EM USD-denominated external debt. This combination of lower funding costs on local and external debt has yet to feed through into faster credit creation and economic activity in the emerging world. With energy supplies becoming more plentiful thanks to the shale revolution, an expansion of capital availability in EMs due to negative interest rates in Europe and Japan could prove to be a big secular stimulus to many emerging economies. True, lower lending standards may encourage more bad loans, but this problem is unlikely to show up in the next 6 to 12 months.

In fact, car sales continue to grow across key EM countries, with India and China posting YoY growth figures of 12 and 10% respectively. In our view, demand could even recover in Russia and Brazil following the tremendous recovery in their currencies, adding further upside pressure on oil demand. Road transportation is not the only area of demand growth in EMs. Petrochemicals, heating, and, of course, passenger air traffic have all continued to expand at a strong pace. In fact, air traffic demand in North America has continued to underperform global growth in recent years, suggesting EM appetite for air travel remains very strong.

The key point to highlight on the demand side is that our expectations are rather modest given the low prices and wide availability of fuels. As we discussed at the start of the piece, we project global demand growth to average 1.2mn b/d next year, down from 1.4mn b/d this year and 1.6mn b/d last year. On a regional basis, we see EM demand being strongest in India, followed by China, and the Middle East, and thus expect liquidity conditions in EMs to remain broadly supportive of continued global oil consumption growth.

To sum up, despite our modest global demand growth projections, we still see the oil market moving into deficit over the next few quarters. Consequently, we retain a constructive view on Brent crude oil prices for 2017 and project prices to average $61/bbl over the course of the year and to touch $70/bbl by the end of 2Q17. To put our projected 2017 balances into context, it is worth highlighting that the last time the market experienced such a deficit was in 2011, a year during which Brent prices increased by $31/bbl. Prices appreciated by 38% on that year compared to the year prior.

To finish off, there are two weather risks ahead worth highlighting. First, a fading El Niño could result in above normal hurricane activity over the next 3 months. NOAA forecasts show a 70% chance of 12-17 named storms this year, of which 2-4 could be major hurricanes and thus create upside risks to petroleum product prices. Most storms tend to be mildly bullish crude and bearish cracks as they temporarily shut down crude production in the Gulf. Occasionally, big storms like in 2005 (Katrina/Rita) and 2008 (Ike/Gustav) come close enough to the shore to shut down more refineries than production, which is rather bearish crude and bullish cracks. In both cases, however, product prices are likely to benefit. Secondly, a La Niña winter may create two way risk for US heating fuel demand. But in the rest of the world, La Niña is clearly associated with a colder than normal winter in pretty much every region except Northern Europe.