The Zhitong Finance App learned that J.P. Morgan's Global Market Strategy Department issued a strategy report for emerging market stocks (EM) on November 11. The title was a question: “What are we talking about when we talk about “proliferation”. The answer is three-tiered: changes in the US political landscape (Democrats swept the midterm elections) removed the geopolitical risk premium that weighed on EM; three mutually reinforcing tailwinds — falling oil prices, falling bond yields, weakening or stabilizing the dollar — opened at the same time; and the true meaning of “diffusion” is that the rise spread from a few AI-weighted stocks to the non-tech part of EM.
What exactly is “diffusion” expanding
The report uses three maps to define the problem. First chart: The contribution of EM's return during the year was highly concentrated in the top five stocks — the index is rising, but only a few. Second chart: Outside of IT, MSCI EM's performance during the year was quite lackluster. Third chart: The trend of EM and other weighted indices compared to MSCI EM. The report says this is “the second round of diffusion trading this year.”
If you connect the three pictures, the logic is clear: EM in 2026 was a narrow rise driven by AI hardware. IT in Taiwan, South Korea, and mainland China supported the index; once AI trading falters, the index will fall, but waiting for weight and small market may outperform. This is the first meaning of “diffusion” — it's not that EM as a whole has risen more, but rather that the increase has spread from a few names to the rest.
The second level of meaning is in the valuation table. The report gives two sets of numbers: EM (excluding IT) is expected to grow at 24% and 11% in 2026 and 2027; developed markets (excluding IT) are 23% and 6.5%; while EM excluding IT's 12-month forward-looking price-earnings ratio is only 10.5 times, which is about one standard deviation below the historical average of 11.3 times. The profit gap has been reversed, and the valuation gap remains the same.
First off: oil prices, established and verifiable
This is the only one of the three that has been fully realized in terms of data.
At the end of October, Brent rushed to about 118 US dollars/barrel due to market concerns that the US would step up military action against Iran after the midterm elections. After the Democratic Party won both houses, the geographical premium began to rebound: Brent fell 5.85% to $92.57 in a single day on November 9, then fell 1.2% to $91.49 on November 10. At one point, it fell below 90 to about $89.95 in the intraday period, the lowest since August 31; WTI received $86.04.
The supply side is also cooperating. According to the survey, OPEC production increased by about 1.38 million b/d to 22.87 million b/d in October, and Saudi Arabia, Kuwait, and Iraq simultaneously increased production; J.P. Morgan's commodity team itself estimated that global oil supply exceeded consumption by about 1.4 million b/d in October, and 1.1 million b/d in September — the original report used this set of figures. The demand side is the opposite: the US Energy Information Administration expects global oil consumption to fall from a record 104.4 million b/d in 2025 to 102.4 million b/d in 2026.
As a result, the report introduced a transmission chain: low oil prices → reduced import inflation → pressure on policy interest rates eased → monetary pressure on importing countries eased → profits in the consumption, retail, automobile, and financial sectors of the oil import market improved. The beneficiary markets named in the report are India, South Africa, China, and South Korea.
The report also acknowledged that this chain is not unidirectional: it is a net negative for oil export markets such as Latin America; however, it added — oil exporters are also worried about inflation and high interest rates, so falling oil prices are not necessarily bad for them at all. Another variable included in the report is El Niño: the report says the intensity is unprecedented; the impact will be felt 3 to 5 quarters after the peak, with Southeast Asia and Latin America being exposed.
Second fine wind: bond yields, but the report appears to have been overturned on the day it was finalized
The report reads: After the Democratic Party swept away and oil prices plummeted, America's long-term yield fell to a low level for several weeks, and Britain and Germany also fell at the same time last Thursday; the bond market is pricing “splitting Congress will bring fiscal restraint and speed up the end of tension in the Middle East”; the possibility of splitting Congress to reduce large-scale tax cuts and heavy spending plans; there is also more supervision over debt ceiling discussions. The conclusion is that low yields increase risk appetite, and EM stock may resume inflows.
This statement was correct at the time it was written, but after the CPI came out in October, the market changed its pricing: the 10-year yield rose by about 8 to 14 basis points in a single day, once hitting 5.025%; the two-year period rose 17 to 20 basis points to 4.74% — 4.755%; and the 30-year period rose to 5.357%. According to CME data, the probability of raising interest rates by 25 basis points in December was 84.2%, and the probability of raising interest rates by 50 basis points rose from about 1% to 11.8% — 13.6%. Federal Reserve Governor Waller stated his support for the December rate hike before the data was released, and said that if the core CPI is 0.3% or above, “one 50 basis point rate hike should be considered”; Dallas Federal Reserve Governor Logan said at least two 25 basis point rate hikes are needed.
It is currently a cycle of interest rate hikes, not a cycle of interest rate cuts. The report describes this risk as “the Fed's interest rate hike exceeds the current market price” — it is not that interest rate cuts are not fast enough, but that interest rate hikes exceed expectations. However, the ten-year yield is currently stuck on the line set by the report: the report said, “A ten-year yield of less than 5% may act as a pressure relief valve.” On November 10, it reached 5.025%, closing at around 5%.
Article 3: The US dollar is an option rather than a benchmark situation
The report's treatment of the dollar was much more careful than the previous two articles, and this is worth pointing out separately. Its argument is based on historical statistics: in the past 20 years, the median MSCI EM (in dollar terms) rose 3.4% for every 1% weakening of the dollar; in these ranges, EM was the best performing region in the world; Peru, Brazil, Turkey, Hungary, the Czech Republic, Indonesia, and South Africa — mostly double deficit economies — showed a median performance of over 25%; in the sector, materials, industry, finance, optional consumption, and real estate outperformed.
The bank said that even if the dollar stabilizes or weakens only moderately, it will be enough to boost EM, especially when EM fundamentals are improving. What the foreign exchange strategy team is doing this week is to reduce overall foreign exchange exposure — reducing both dollar bulls and cyclical foreign exchange. The reason is that the AI-led stock market sell-off after the midterm elections is challenging the pillar of “American exceptionalism,” and the exchange rate will follow the stock market in the next few days rather than activity data.
Therefore, the real weight of the three fine winds is: one has been realized (oil price), one has been overturned by today's data (yield), and one is an option rather than a benchmark (US dollar). The report juxtaposed them as “three mutually reinforcing”. This statement is stronger than actual evidence.
So what support is left for “diffusion”
The two remaining supports are not at the macro level.
The first one is funding. The report warns that in the two months before the Middle East conflict broke out (January-February 2026), foreign subscriptions for EM equity funds were at an all-time high, and have begun to pick up again in recent weeks. India is a verifiable example: foreign institutional investors (FII) have returned around $7.5 billion since mid-October after experiencing an unprecedented sell-off during the year.
The second is the return of active management. The logic is: after the market becomes more discerning, active managers can avoid overvalued sectors and concentration risks, and target areas of continuous growth to generate excess profits, thus attracting more active allocation. This one is more like inference than evidence — the support is just a 12-month chart of EM Active Fund's performance versus MSCI EM, with no forward-looking funding predictions.
Four markets, four sets of logic
The “Market Implications” section of the report is organized by region, and the four sets of logic are independent of each other.
China: May benefit from short back-up and capital rotation — Chinese stocks were previously used as a source of short funding to prolong AI. The report argues that the Chinese stock market provides decentralization of global AI transactions, and that the localization of AI in China has its own trajectory and may be less dependent on US AI capital expenditure. Its Chinese strategy team believes that as valuations recover from the bottom, there is still room for a 10% to 15% rebound, and additional policy incentives may be extended further.
India: Risk appetite is gradually improving. The recovery in October was due to relative defensiveness against global AI sell-offs and third-quarter results exceeding expectations. Crude oil imports are around 85% to 90% dependent, and Middle East peace is a big plus — smaller import bills, more stable rupees, lower inflation, and higher profit margins.
South Africa: The report's top pick in CEEMEA, maintaining overallocation throughout the conflict. Comparisons between the two groups: South Africa and Brazil are commodity exporters, high forex beta, and MSCI EM weights are similar (about 3% — 4%); South Africa's bond yields have declined by about 350 basis points since the June 2024 national election, but the stock market has not been meaningfully revalued. Recent local elections have improved the prospects of the ruling coalition (DA); the sector favors banks and gold stocks.
Mexico: The report sees it as one of the biggest global beneficiaries of America's “blue sweep” — tariffs may be more moderate, and USMCA negotiations may be more substantial. The peso has depreciated more than 10% in two months, and the central bank has ended the easing cycle and raised the threshold for further depreciation; the allocation of foreign capital is very low, which is an alternative to “proliferation” in Latin America (considering Brazil's macropolitical debate).
AI: Still bullish, but admits it's a “reset in the cycle”
The report didn't abandon AI. The three fundamental reasons it gave are worth fully recording: the unit economy, the law of scaling, and token demand growth are still strong; even in a double-digit growth environment, the valuation has been revised to a level that suggests “profits are about to be lowered,” and it believes this is unlikely to happen; the typical peak signs of a cycle of inventory accumulation, shortened delivery times, and rapid expansion of production capacity are basically missing.
Based on this, it characterized the current pullback as a “reset in one cycle and another deleveraging event within the AI supercycle,” which is more optimistic than the “end of cycle” pricing in the market. The risk is to wait and see the market. An additional variable brought about by the midterm elections is a policy ceiling for data center construction: the blue sweep may limit the capital expenditure curve, but the bank's US strategy team believes that the construction trajectory is intact, the key risk is delay rather than cancellation, and the focus will shift to more supportive jurisdictions.
Bottom line: The denominator side of this arithmetic question is still in someone else's hands
After breaking up this report, its real contribution was not the term “three tailwinds,” but rather turned “proliferation” from a narrative to a testable arithmetic problem: the profit growth rate of EM (excluding IT) has surpassed developed markets (excluding IT), yet the valuation is still one standard deviation below the historical average. This rule does not depend on elections, oil prices, or the dollar — it is the only argument that can be established independently under any macroeconomic circumstances.
But it also revealed its own vulnerability: the numerator (profit) of this arithmetic problem is EM's own business, but the denominator (discount rate) is in the hands of the Federal Reserve. The report itself set a threshold of “less than 5% over a ten-year period,” and this threshold was touched upon on November 10; it listed the risk that “the Federal Reserve raised interest rates more than the market price,” and the market just switched from “to add or not” to “how much to add.”
So what you really need to focus on is not EM's profit statement, but the December interest rate meeting. If interest rate hikes are implemented and the long-term end is stable below 5%, “diffusion” has the denominator it needs; if 50 basis points become a reality and 5% over a ten-year period, then the smooth wind brought by oil prices is probably only enough to offset the rise in financing costs — the first smooth wind and the second headwind of the report will cancel each other out; the only remaining leg is valuation discount.