Recently, during our roadshow, we’ve deeply felt a significant shift in investors’ understanding of environmental protection-driven production restrictions. This shift has almost entirely taken place over the past two weeks. At the outset of our roadshow, which began the week before last (starting September 11), investors virtually equated environmental protection-driven production restrictions with rising industrial commodity prices—meaning that stricter environmental restrictions inevitably lead to price hikes, and the extent to which these restrictions are implemented directly determines the magnitude of those price increases. Consequently, investors’ primary concern has been precisely how effective the implementation of these environmental restrictions will be—and whether even tougher policies might be introduced.

　　As the roadshow progressed into its latter half, despite the continued strict implementation of environmental production restrictions, commodity futures prices began to fall rapidly. Investors gradually realized that these environmental restrictions were actually producing a “double whammy” effect—simultaneously hitting both supply and demand. We’ve already highlighted three examples of this supply-and-demand double whammy: the shutdown of profile processing plants in the Tangshan region, the closure of aluminum profile processing plants in Shandong, and the suspension of operations at auto parts processing facilities in Shanghai. All of these developments have delivered a fairly substantial negative impact on raw material prices—prices for which earlier there had been high expectations. (These include steel billets, electrolytic aluminum, and steel plates.) This situation has resonated strongly among many clients; several enthusiastic customers even provided us with additional evidence illustrating the supply-and-demand double whammy. For instance, some mentioned that steel traders now feel compelled to offer “value-added services” to their customers—selling steel products while simultaneously helping them source complementary parts and components. Others questioned the logic behind the recent price hikes, pointing out that inventory levels are currently quite low—not only among traders but also among steel-using enterprises. With high profits at steel mills, traders and steel users are more likely to form a united front, refusing to accept the relatively high steel prices and forcing mills to make “genuine” concessions. Another customer pointed out that just before this year’s National Day holiday, steel mills showed little enthusiasm for stocking up. In previous years, downstream steel users typically started stockpiling two weeks ahead of the holiday—in other words, beginning in mid-September—but this year, the willingness to stock up was much weaker. The steel mills’ determination to hold prices and withhold supplies stood in stark contrast to the hesitation and reluctance of traders and steel users, who were wary of further price increases. From this perspective, we’re not optimistic about prices after the National Day holiday either. The reason why steel users and traders dare to take a tough stance against the mills is precisely because downstream demand remains relatively subdued. Furthermore, some customers cited evidence of halted earthmoving operations at construction sites, suggesting that a decline in demand for long products is inevitable.

　　We are grateful to our customers for their support of our work. Their selfless sharing has enhanced and deepened our understanding and reflection on the issue of steel prices. If low inventory levels have already become evidence of a bearish outlook on steel prices, it clearly underscores the pessimistic sentiment in the market. The logic that “environmental production restrictions are a double-edged sword” has now taken root in people’s minds. Moreover, the relatively weak economic data for July and August have further intensified market pessimism about future demand. One customer put forward this view: “Low inventory and a negative supply-demand gap can only be a necessary condition for price increases—not a sufficient one. Only when demand remains strong enough to offset the negative supply-demand gap can prices actually rise. Right now, demand is clearly weakening, so it’s unreasonable for steel companies to hold such high profit margins.” Recently, we’ve observed that even after spot steel prices were adjusted, trading activity remains sluggish, and market sentiment is overwhelmingly cautious. As we’ve mentioned before, during previous steel price hikes, too, market sentiment was similarly cautious. This is precisely a characteristic of a situation marked by weak overall demand. Whether prices are rising or falling, steel-using enterprises remain largely indifferent—unaffected by the supply-demand gap. This customer’s viewpoint is indeed well-founded.

　　Moreover, we previously mentioned a point: when a particular commodity experiences intense speculative hype, its price does not settle into a stable equilibrium but instead tends to diverge. Let’s briefly consider this question: What significant differences exist in the supply and demand for steel when the profit per ton is 1,000 yuan versus 500 yuan? At 500 yuan per ton, companies’ willingness to supply indeed weakens somewhat. However, under conditions of production restrictions, the primary factor influencing supply has already shifted to the restriction policies themselves. When prices fall, if companies hold a pessimistic outlook, they may actually accelerate sales, striving to realize revenue while profits are still relatively good. Thus, as prices decline, supply may not necessarily contract. On the demand side, on the one hand, production restrictions remain the key variable affecting demand; on the other hand, steel demand hardly shows a clear inverse relationship with price. Construction sites and automobile manufacturers cannot simply increase their steel usage just because steel prices have fallen. Therefore, the supply and demand at 1,000 yuan and 500 yuan per ton of steel might not differ significantly at all. Given this, how can we possibly determine whether the current supply-demand gap is more reasonably matched to a profit of 1,000 yuan per ton or to a profit of 500 yuan per ton?

　　We’d like to share with you some preliminary insights for your reference: A supply-demand gap does not prove that a particular price level is justified; rather, it’s just one of several factors that determine the direction of price movements—but by no means the only one. If we were to conclude that steel prices could rise simply because there’s a supply-demand gap, then no matter whether the profit per ton of steel is 300, 500, 1,000, or 2,000 yuan, the conclusion that prices will rise would hold true—and that would be a conclusion that defies common sense. To make accurate price judgments, we also need to take into account the following points: First, the strength or weakness of overall demand. As we’ve mentioned earlier in our discussion of customer perspectives, if overall demand declines too sharply and too quickly due to production restrictions, even if a supply-demand gap still exists, prices will find it difficult to rally. Pessimistic expectations about overall demand on both the supply and demand sides will influence their bargaining attitudes: suppliers will want to sell their goods as soon as possible, while buyers will seek to delay taking delivery.

　　Second, let’s talk about the impact of the futures market. After our recent discussions, we’ve all generally noticed that quantitative trend traders are wielding increasing influence in the commodity and government bond futures markets. As we mentioned in our previous commodity commentary, there’s a close linkage between commodities and government bond futures. If we plot the intraday (minute-level) data for both government bond futures and the main contract for threaded steel together, we’ll find that their movements are strikingly inverse. It’s highly likely that trend traders are “betting on both sides” in the market. Investors should pay attention to this phenomenon: recently, the implied repo rate (IRR) for the main government bond futures contract has been exceptionally high—so high that arbitrage trades with a positive carry have virtually guaranteed profits (though not necessarily so when compared to financing costs, but considering the value of the futures contracts themselves, the odds of winning are very favorable). Meanwhile, open interest continues to rise rapidly. We believe that genuine bond investors have absolutely no intention of taking the opposite side in these arbitrage trades. Such high futures prices are most probably being driven up by trend traders. On Friday, government bond futures rallied once again, moving in the opposite direction from commodities. All these developments indicate that the participation of quantitative trend traders is intensifying the previous market linkage, making price trends more likely to self-reinforce and gain momentum—and consequently affecting the spot market as well. The sharp decline in commodity futures on Friday directly led to a substantial drop in today’s (Saturday’s) spot prices, and this could even spill over to influence Monday’s opening futures prices.

　　Although we have remained cautious throughout this price cycle, the rapid decline in steel prices has still slightly exceeded our expectations—indicating that we somewhat underestimated the strength of trends in the futures market. Given the current situation, while we wouldn’t dare predict that futures prices will continue to plunge sharply, it’s now extremely difficult for the market to stage a major rebound in the short term. Spot prices have also begun to show clear signs of easing, providing relatively solid support for the bond market. As we discussed in our previous report, commodities are gradually shifting from predominantly negative factors toward more positive ones. This shift reflects investors’ growing recognition of a potential slowdown in overall future demand—a point that warrants our attention. For instance, despite currently robust financial data, many investors have expressed underlying concerns. Since the beginning of this year, social financing has been largely driven by loans; yet both residential loans and medium- and long-term corporate loans logically face the risk of slowing down (due to the lagged effects of real estate sales, tighter regulations on consumer loans, and shrinking local-government investment and financing). As a result, there’s an emerging possibility of self-sustaining improvement in liquidity conditions—something that may not have received much attention from the market previously. We believe these developments represent relatively positive shifts in the market.

　　Looking ahead, we believe that short-term commodity prices are likely to remain in a weak, volatile range, pulling the PPI down from its recent peak. However, once futures prices stabilize after an initial sharp decline, inventory levels and supply-demand gaps will once again become the primary drivers of price movements. Moreover, differentiation among various commodities may widen once again—these are all issues that warrant our continued attention. In the latter part of this weekly report, we have specifically summarized the supply-and-demand conditions for several key industrial products (steel, coke, cement, aluminum, and iron ore) for investors’ reference. Due to space limitations, we provide only brief descriptions of these supply-and-demand dynamics without going into excessive detail. For specific data, please feel free to contact us to request further information.

　　Analysis of Supply and Demand for Key Industrial Products During the Environmental Protection Production Restriction Season

　　In March, the Ministry of Environmental Protection, together with other government departments and six provinces (and municipalities), jointly issued the “2017 Work Plan for Air Pollution Prevention in the Beijing-Tianjin-Hebei Region and Surrounding Areas.” The plan covers 26 cities along the atmospheric pollution transmission corridors in the Beijing-Tianjin-Hebei region, plus the two directly-administered municipalities of Beijing and Tianjin—hence the term “2+26” cities. Specifically, the “2+26” cities include Beijing, Tianjin, Shijiazhuang, Tangshan, Langfang, Baoding, Cangzhou, Hengshui, Xingtai, and Handan in Hebei Province; Taiyuan, Yangquan, Changzhi, and Jincheng in Shanxi Province; Jinan, Zibo, Jining, Dezhou, Liaocheng, Binzhou, and Heze in Shandong Province; and Zhengzhou, Kaifeng, Anyang, Hebi, Xinxiang, Jiaozuo, and Puyang in Henan Province. On August 21, the Ministry of Environmental Protection again released the “Comprehensive Action Plan for the Prevention and Control of Atmospheric Pollution during the Autumn and Winter of 2017-2018 in the Beijing-Tianjin-Hebei Region and Surrounding Areas,” requiring local governments involved to issue more specific action plans by the end of September.

　　Steel

　　In the steel sector, according to the production restriction plan released in March, key cities are required to intensify efforts to limit steel production. Localities will implement a tiered management system for steel enterprises, formulating staggered production restriction and shutdown plans based on their pollution emission performance levels. In key cities such as Shijiazhuang, Tangshan, Handan, and Anyang, steel production capacity will be restricted by 50% during the heating season, with the reduction calculated based on blast furnace production capacity and verified using actual electricity consumption data from the enterprises. Subsequently, in the specific production restriction plans issued by various cities, Tianjin, Zibo, Changzhi, Jiaozuo, and Jincheng have explicitly set a 50% reduction target for steel production. Overall, among the “2+26” cities, nine cities have explicitly set steel production restrictions exceeding 50%.

　　There are various interpretations in the market regarding the 50% reduction in steel production capacity. According to data from MySteel, the current utilization rate of steel blast furnaces stands at 84%. If the 50% capacity reduction takes into account blast furnaces that are currently underutilized, the actual reduction in output could be around 30%, which is significantly milder than the stated 50% reduction. Our understanding, based on verification using enterprises’ actual electricity consumption data as stipulated in the plan, is that the production cuts should be aligned with the approved compliant capacity. Therefore, we anticipate that production during the heating season will also decline by approximately 50%.

　　The impact of production restrictions can be considered either from the perspective of production capacity or from the perspective of output. However, given that provinces have been continuously working to reduce overcapacity since last year, we anticipate that using output data for calculation would be more reasonable.

　　Since the heating season restrictions primarily affect blast furnace ironmaking capacity, electric arc furnace steel produced through short-process routes has not been affected. In 2015, China’s electric arc furnace steel accounted for approximately 6.1% of total steel production, with the remainder being converter steel—both falling within the scope of these production restrictions.

　　According to the production data we obtained from the 2016 Statistical Bulletins on National Economic and Social Development released by various cities, the combined output of pig iron in the “2+26” cities totaled 164.3 million tons, crude steel output totaled 194.91 million tons, and steel product output totaled 400.72 million tons. Among these, for the nine cities that have explicitly announced a 50% reduction in steel production, the combined output of pig iron, crude steel, and steel products amounted to 151.37 million tons, 170.80 million tons, and 306.20 million tons, respectively (Figure 1). Based on the production restriction periods announced by some provinces and cities, we estimate that this year’s heating season will still last from November 15, 2017, to March 15, 2018—nearly four months. Monthly steel production has shown little variation and no obvious seasonal effects; therefore, we can assume that heating-season production accounts for approximately one-third of the annual total. From this, we can calculate that if all “2+26” cities were to implement a 50% production cut, the impact on output would be 62.71 million tons. However, if we consider only the nine cities that have already explicitly announced a 50% reduction, the impact on steel production would be 47.92 million tons. Given that the total steel production for 2016 was 1.138 billion tons, and the estimated heating-season production is 3.79 billion tons, the production cuts would affect heating-season output by 13% to 17%.

　　On the demand side, in addition to the steel industry, the construction and building materials sectors are also subject to production restrictions. According to the action plan released on August 21, during the heating season, all cement, brick and tile kilns, ceramics, gypsum board, and other building materials industries will halt production—except for those essential for ensuring people's livelihoods such as heating supply. Localities should, based on the specific industrial characteristics of their local building materials sectors, propose broader-ranging staggered-production requirements. According to data from the Metallurgical and Construction Research Institute, construction steel accounts for 55.4% of domestic steel consumption, making it the dominant segment of domestic steel demand. Therefore, the current production restrictions on building materials will also affect the demand for construction steel. According to our statistical data, the total output value of the construction industry in the “2+26” cities in 2016 was 1.64 trillion yuan (excluding production from other provinces), accounting for 8.4% of the nation’s total construction industry output value. According to data from the Metallurgical and Construction Research Institute, national construction steel consumption in 2015 was 350 million tons. Given that steel production in 2016 did not significantly deviate from the 2015 level, we assume that 2016 consumption remained at 350 million tons. Consequently, the planned restrictions on the building materials sector are expected to reduce steel demand by 104.6 billion tons.

　　As shown in the analysis above, production restrictions during the heating season will lead to a decrease in steel supply ranging from 479.2 billion to 627.1 billion tons. Meanwhile, production restrictions downstream in the steel industry will affect demand by 104.6 billion tons, leaving a certain supply-demand gap during the heating season. However, given the significant response of steel prices to these measures previously (Figure 2), coupled with the fact that many steel mills are already proactively stocking up on raw materials such as sintered ore (Figure 3), it is expected that the actual supply-demand gap will be smaller than estimated. Consequently, it will become increasingly difficult for steel prices to continue rising sharply, and they are likely to adopt a more volatile pattern instead.

　　Coke

　　Regarding coke production, according to the action plan released on August 21, the requirements for the coking industry are as follows: From October 1, 2017, to March 31, 2018, coking enterprises must extend their coke-quenching time to more than 36 hours. Coking enterprises located in built-up urban areas must extend this time to more than 48 hours. Thus, coking enterprises will face production restrictions for a period of six months, and the restricted areas remain the “2+26” cities. Given that the average coke-quenching time is currently 24 hours, extending the quenching time to more than 36 hours would result in a production reduction of 33%; while extending it to more than 48 hours would lead to a 50% reduction. We assume an average production restriction rate of 40%.

　　According to our statistical data, the combined coke production of the “2+26” cities totaled 77.2 million tons. Based on an average production restriction rate of 40%, we can estimate that the total supply of coke during the autumn and winter seasons in the “2+26” region will decrease by 15.44 million tons (Figure 4).

　　Regarding coke demand, since most coke is used in pig iron production, a decline in steel output will affect pig iron production, thereby impacting coke demand. In 2016, pig iron production was 70,000 tons, while steel production reached 1.138 million tons, with pig iron accounting for approximately 62% of total steel output (with scrap steel serving as a supplementary raw material). Based on the calculations mentioned above, steel production during the heating season is expected to decline by 47.92 million to 62.71 million tons, implying that pig iron demand will drop by 29.71 million to 38.88 million tons. Given that 0.5 tons of coke are required to produce 1 ton of pig iron, coke demand will therefore decrease by 14.85 million to 19.44 million tons.

　　From this, we can see that the decline in coke supply is relatively balanced with the decline in demand. However, the timing of these declines is not synchronized: the decline in coke supply spans from October 1 to March 31 of the following year, while the decline in demand is concentrated during the heating season. This could lead to stronger demand for coke from October 1 to November 15 and from March 15 to March 31. As a result, there is a possibility of phased price increases for coke; yet overall, the market may still remain in a volatile pattern. If prices of other products in the black industry chain fall across the board, coke prices will likely find it difficult to remain unaffected.

　　Cement

　　According to the action plan released on August 21, during the heating season, the building materials industry—including cement (including special cements but excluding grinding stations), brick and tile kilns (excluding those fueled by natural gas), ceramics (excluding those fueled by natural gas), glass wool (excluding those fueled by natural gas), rock wool (excluding those produced using electric furnaces), and gypsum boards—will all suspend production. According to our statistical data, the total cement production in the “2+26” cities amounts to 311.64 million tons (Figure 5), accounting for 12% of the nation’s total cement output. If all “2+26” cities were to halt production entirely during the heating season, cement output is expected to decline by approximately 10,000 tons. Given that cement storage periods should not be excessively long and that enterprise inventories are relatively low, it is anticipated that certain regions will experience a supply-demand gap during the heating season, potentially creating room for structural price increases in cement.

　　Ore

　　Based on our previous estimates of changes in pig iron demand, it is relatively easy to infer a corresponding reduction in ore consumption. Roughly speaking, 1.5 tons of ore correspond to 1 ton of pig iron. Therefore, the reduction in ore demand ranges from 44.57 million tons to 58.32 million tons, accounting for approximately 5% of the annual import volume. Moreover, current iron ore inventory levels remain relatively high (Figure 6). At the same time, the price spread between high-grade and low-grade ores has begun to narrow somewhat, possibly indicating that the previous strong production incentive—driven by the desire to maximize output by using as much high-grade ore as possible—is easing. As a result, the overall oversupply situation in the iron ore market remains unchanged, making it difficult for prices to rise further. Instead, prices are likely to fluctuate weakly.

　　Electrolytic aluminum

　　In March of this year, the Ministry of Environmental Protection and other relevant departments issued a notice imposing production restrictions during the heating season on certain high-pollution industries in this region: Electrolytic aluminum enterprises are required to reduce production by more than 30% (calculated based on electrolytic cells); alumina enterprises are required to reduce production by around 30% (calculated based on production lines); and anode carbon enterprises are required to reduce production by more than 50% (calculated based on production lines; if their emissions do not meet the special emission limits, they must halt production entirely).

　　Regarding the specific impact assessment: During the heating season, production restrictions imposed on the “2+26” cities will affect approximately 1 million tons of output, primarily impacting enterprises with production lines located in the Beijing-Tianjin-Hebei region and its surrounding areas, such as Hongqiao and Xinfu. According to data from Aladdin, the “2+26” cities account for an electrolytic aluminum operating capacity of 13.18 million tons (representing about 30% of the nation’s total capacity). As per regulations, production restrictions are mandated at 30% based on the number of electrolytic cells that are shut down. Given that the statutory heating period in northern China lasts 120 days—from November 15 to March 15 of the following year—this period will serve as the benchmark for implementing the production restrictions. Based on the current industry average capacity utilization rate of 87%, it is estimated that the total affected capacity during the 2017-2018 heating season will amount to roughly 4 million tons, equivalent to an output reduction of approximately 1.1 million tons.

　　Under such expectations, aluminum prices had risen rapidly in the earlier period. However, this rise in aluminum prices has also boosted corporate production profits and triggered a rapid increase in electrolytic aluminum inventories. Meanwhile, the continued rise in the price of prebaked anodes—their key raw material—has highlighted the tight supply situation recently caused by manufacturers’ strong willingness to ramp up production. Currently, electrolytic aluminum inventories have surged to historic highs, rising from 260,000 tons in July 2016 to nearly 1.2 million tons, reaching a total of 1.47 million tons. This means that although, from a marginal supply-and-demand perspective, electrolytic aluminum is facing a supply shortage in the fourth quarter of this year and the first quarter of next year (with quarterly production of around 9 million tons), at this stage, the industry’s high inventory levels can essentially offset the supply-demand gap during periods of production restrictions (Figure 7). Consequently, the room for further substantial increases in aluminum prices may be relatively limited going forward, and as we enter the heating-season production restriction period, the price center may not necessarily continue to rise compared to current levels. Nevertheless, electrolytic aluminum still faces certain uncertainties, which could lead to phased price increases: First, the timing and intensity of environmental inspections and enforcement; second, the potential shortage of prebaked anodes—the raw material used in aluminum production—if environmental production restrictions require anode carbon enterprises to cut production by more than 50%.

　　First, there is some uncertainty regarding the timing and intensity of environmental inspection enforcement. On the one hand, a sudden, large-scale shutdown of electrolytic cells can significantly impact the power grid; therefore, companies typically plan to shut down their cells in phases, meaning that aluminum producers will begin shutting down earlier and it will take some time before they fully resume production. On the other hand, on September 18, 2017, the Jiaozuo City Environmental Protection Task Force issued an “Urgent Notice on Further Strengthening Air Pollution Control Measures.” The notice stated that, starting from 0:00 on September 19, Jiaozuo Wanfang Aluminum Co., Ltd. and Chinalco Zhongzhou Aluminum Co., Ltd. would implement staggered production schedules ahead of schedule, limiting output and emissions by more than 30% until March 15, 2018. This means that in certain regions, the timing of these production restrictions has been brought forward, which will have a phased impact on subsequent supply and market expectations.

　　Second, in the case of primary aluminum production, electricity costs and alumina account for the two largest components of production costs. Among electricity costs, prebaked anodes require coal pitch and petroleum coke—two raw materials that, due to ongoing environmental pressures, may soon face supply shortages. Considering the nation’s total capacity for anode carbon products, which stands at approximately 26 million tons, the operating rate, under environmental restrictions, may fall below 70%. This limited supply support for primary aluminum production could constrain subsequent aluminum production. (Figure 8)

　　Overall, aluminum electrolysis and alumina production capacities are expected to continue shrinking in the coming period, while capacity utilization rates are likely to keep rebounding. We anticipate that aluminum electrolysis prices will remain volatile yet generally strong throughout the year.