Aluminum Rally Will Be Foiled

Amid a substantial rally, aluminum spot prices have reached their highest level since February 2013. Prices have moved higher because of better-than-expected aluminum demand as well as the perceived benefits of capacity reductions in China.

We believe investors have become overly enthusiastic on both counts. Instead, we forecast a significant deceleration in aluminum demand growth and anticipate that the impact of capacity cuts will prove far overstated. Accordingly, we forecast a long-term aluminum price of only $1,475 per tonne (in real terms), nearly 30% below current levels.

Our long-term midcycle price forecast, to be achieved by 2021, sits 18% below the Metal Bulletin consensus outlook of $1,790 per tonne (in real terms), which is based on the published figure just above $2,000 (in nominal terms). A price decline of this magnitude would have a substantial impact on the share prices of

The primary driver behind our admittedly bearish outlook for aluminum prices is our below-consensus forecast for Chinese fixed-asset investment. As China’s economic growth model becomes more consumption-driven at the expense of the heavy investment contributions witnessed over the past two decades, we anticipate that gross capital formation growth will be effectively zero over the next five years. This forecast compares with consensus forecasts of 3%-5%. Although the delta in percentage terms might appear modest, the implications for broader industrial metal prices would be massive, given China’s outsize impact on global commodity demand. Accordingly, we also maintain below-consensus price forecasts for steel, iron ore, metallurgical coal, and copper.

Although our outlook for limited investment growth is based on structural factors stemming from broad changes to China’s economy, the waning effects of recent stimulus measures will also serve as a headwind. China’s most recent economic stimulus package, introduced over the course of 2015 and 2016, targeted $1.1 trillion of infrastructure spending across 300 projects. Higher infrastructure spending has, in turn, catalyzed industrial metals demand, driving higher prices. However, we view stimulus efforts as a short-term fix for a more substantial long-term issue. Although these efforts have buoyed fixed-asset investment growth for a number of quarters, fixed-asset investment growth has again decelerated in recent months. Stimulus packages of this magnitude can’t be ruled out in the future, if the central government continues to believe that they will help usher a smoother transition for the broader economy. However, we also observe that each iteration of stimulus is generating less and less growth; in other words, the return profile for each incremental unit of stimulus currency spent will continue to decrease. In part, this is because more and more capital is being used to pay down bad debts rather than being invested in projects that might otherwise generate attractive returns. This decreases the likelihood that the central government will undertake future spending programs of this magnitude, and if it does, the incremental impact on industrial metal demand will be lower for each currency unit spent.

Sticking with the demand side of the equation, we’re often asked if a boom in India’s commodity demand will offset a bust in China. We forecast that India is still roughly two decades away from becoming a major global player with regard to aluminum consumption (as defined by reaching a 10% global consumption share). Although India’s consumption growth potential in the coming decades is unmatched in absolute terms, the country is simply too early in its economic development (as measured by GDP per capita) to become a major global aluminum consumer any sooner. In 2016, India accounted for only 3.1% of global consumption with GDP per capita of about $6,000. Typically, aluminum consumption per capita doesn’t accelerate meaningfully until a country surpasses the $10,000 GDP per capita threshold. India is unlikely to reach this milestone until the middle of next decade. Even at that point, however, it will be some time before the country becomes a major global player in terms of its aluminum consumption share.

On the supply side, many analysts have cited Chinese aluminum capacity cuts as a key driver of higher prices. However, assuming that the government’s targeted capacity reductions are achieved, the ultimate impact on aluminum prices will be more than offset by the continued addition of new, low-cost capacity. Based on Harbor Aluminum data, 6.8 million tonnes per year of new capacity will be added by the end of 2018, confirmed restarts of capacity previously idled will be 1.5 million tonnes per year, and production cuts will total only 1.2 million tonnes per year. Newly added capacity will almost exclusively fall on the low end of the cost curve, displacing existing facilities further to the right side of the cost curve and weighing on the market-clearing aluminum prices. China’s aluminum industry will probably remain in a surplus position through 2018, if not longer. We believe the perceived effect of capacity cuts has been severely overstated.

Yet another headwind facing the global aluminum industry is the fact that world inventories reached an all-time high in June at 17.2 million tonnes based on Harbor Aluminum estimates. Although global visible inventories have declined in recent years, invisible or “stealth” inventories have continued to climb, totaling an estimated 11.6 million tonnes at June-end. As aluminum prices fall and inventories fall to more “normal” historical levels, the associated unwinding will displace primary aluminum supply. Based on 2016 global consumption, current inventory levels represent 105 days of consumption, far above our “normalized” assumption of 70 days. Although the long-term impact of a gradual inventory drawdown will be far smaller than other factors cited above, it will still prove material.

Finally, with the aluminum premium bubble having popped, we contend that premiums will remain lower for longer, particularly as inventories unwind. The Midwest U.S. aluminum premium peaked at $535 per tonne in February 2015 but has averaged just below $200 per tonne year to date. We forecast a long-term midcycle premium of only $170 per tonne (in real terms), consistent with the current premium of $167 per tonne. Therefore, aluminum smelters won’t benefit from higher premiums as aluminum prices grind lower.

Our negative outlook for aluminum prices (and in turn alumina and bauxite prices) spells trouble for companies under our coverage that operate in the aluminum supply chain. This includes Alcoa, Norsk Hydro, Chalco, and Alumina. Currently, these four companies are all trading well above our fair value estimates, as market prices seem to imply that elevated aluminum prices will be sustained. Instead, we expect all four companies to suffer material margin contraction as the aluminum cost curve flattens, aluminum demand wanes, and the market-clearing aluminum price falls. Of these companies, Norsk Hydro is trading at the lowest price/fair value estimate, as it also consumes large volumes of aluminum for the production of finished aluminum products. High-cost Chinese state-owned operator Chalco offers the worst value by far, as reflected by a share price more than 6 times our fair value estimate.