Global markets are undergoing an adjustment triggered by external shocks, with investors facing two core questions: when will this correction bottom out, and is the current macroeconomic environment replaying the stagflation nightmare of the 1970s? A relatively optimistic but conditional assessment was provided in the latest "Flow Show" report from Bank of America Merrill Lynch on March 7th: signals for the end of the correction are emerging but are not yet fully in place. Furthermore, the 2020s are more likely to head towards an inflationary boom rather than a stagflationary collapse—provided geopolitical tensions do not worsen further. According to analysis by strategist Michael Hartnett's team at Bank of America Merrill Lynch, this correction was triggered by a combination of external shocks and excessive optimism. Signs are appearing that some "oversold" assets are bottoming, but oil prices and the US dollar have not yet given clear reversal signals, and the S&P 500 has not undergone sufficient price clearing, such as falling below 6600 points. Simultaneously, the Bank of America Merrill Lynch Bull & Bear Indicator remains elevated at 9.2, in extreme bullish territory, indicating that market sentiment has not truly cooled, thus limiting the potential for a rebound.
News from NVIDIA also caused market tremors: NVIDIA stated that a previously announced $100 billion investment in OpenAI is "not in the plans," and the current $30 billion financing arrangement might be the upper limit. This statement is seen as a potential signal that the exponential growth in AI capital expenditure might be slowing, an effect that cannot be ignored for tech bonds and the software sector. When will the correction end? Four conditions, two currently met. Bank of America Merrill Lynch believes that a market correction triggered by external shocks against a backdrop of excessive optimism typically requires four conditions to be met before it can be declared over:
First, "oversold" assets bottom out (software, MAGS, private credit, bank loans, Bitcoin); Second, "overbought" assets complete their selling (gold, semiconductors, metals, emerging markets, Europe, bank stocks); Third, "safe-haven assets" lose buying support (oil prices and the US dollar); Fourth, genuine price clearing occurs.
Currently, the first two conditions are showing initial signs. Fund flow data supports this view: this week saw the largest weekly outflow from gold since October 2025 ($1.8 billion), while the energy sector recorded its largest weekly inflow on record ($7 billion), indicating investors are "chasing" into previously overbought sectors. However, oil prices and the US dollar have not shown significant declines, and the S&P 500 has not experienced a full price washout.
Bank of America Merrill Lynch clearly states that a significant rebound should not be expected until the US dollar's trend becomes clearer. The US Dollar Index is the best barometer of global liquidity—if the dollar decisively breaks above 100, it would signify a deepening of the "peak liquidity" theme, further compressing expectations for 2026 interest rate cuts (the market's probability of a Fed rate cut on June 17th has dropped from 100% on January 1st to 37%), and could trigger yield curve flattening and an inflationary oil price shock. Looking at fund flows, US equities experienced their largest outflow in six weeks this week ($13.9 billion), while Japanese equities recorded their largest weekly inflow since October 2025 ($4.2 billion). Korean equity flows were extremely volatile, setting a record for the largest single-day inflow on March 2nd ($6.1 billion), followed by a record single-day outflow on March 4th ($4.7 billion). Will the 2020s replay the 1970s stagflation script? This is one of the most debated macro narratives in the current market. Bank of America Merrill Lynch's stance is that the 1970s are the closest historical analogue for the 2020s, but they are not entirely equivalent. Under the baseline scenario, the 2020s are more likely to head towards an inflationary boom rather than a stagflationary collapse. The logic chain supporting an inflationary boom is clear: political populism (non-establishment party vote share in UK elections surging from 27% in 2024 to 69% in 2026), tariffs and immigration policies reversing globalization, excessive fiscal expansion, Federal Reserve policy compromise, and asset and wealth inflation driven by a "too big to fail" stock market.
These factors collectively create inflationary pressures, but government intervention will suppress the rise in bond yields, ultimately manifesting as a weaker US dollar rather than a sharp increase in long-term rates. In this scenario, commodities, real assets, international equities, and small-cap stocks are the primary beneficiaries. However, the history of the 1970s remains a cautionary tale. Bank of America Merrill Lynch outlines the complete sequence of that period:
From 1970 to 1972, the Nixon administration engineered a boom through aggressive fiscal and monetary easing, with stocks surging over 60%; From 1973 to 1974,失控 inflation combined with an oil shock led to a 45% stock market crash; From 1975 to 1976, after the first wave of inflation receded, assets rebounded, with small-caps and value stocks replacing the "Nifty Fifty" as the new leaders; From 1977 to 1980, the Iranian Revolution triggered a second wave of inflation, stocks fell another 26%, until the Volcker shock finally ended the cycle.
Mapping this to the present, Bank of America Merrill Lynch believes the key variable is the situation with Iran. If the conflict is brief, with oil prices remaining below $90/barrel, the inflationary boom narrative holds, and commodities, emerging markets, and small-cap stocks would benefit once a dollar bear market resumes. If the conflict prolongs (strait of Hormuz blockade, Iranian attacks on regional oil infrastructure), pushing oil prices above $100-$120/barrel, asset allocation would shift towards oil, the US dollar, US tech, and global defense, while energy-import dependent markets like Japan, Korea, and Europe would face the heaviest pressure. Looking at the "mosaic" of asset performance during the 1970s stagflation cycle, gold and commodities ranked near the top of the return charts almost every year, while stocks and bonds delivered mixed results. This historical pattern is already reflecting in current markets—year-to-date in 2026, oil is up 30%, gold has risen 18.3%, commodities overall are up 22.6%, while the S&P 500 has gained a mere 0.3%, and Bitcoin has fallen over 16%.
NVIDIA Abandons $100 Billion Deal, AI Capex Narrative Shows Cracks NVIDIA's statement this week that the previously announced $100 billion investment in OpenAI is "not in the plans," and that the current $30 billion financing might be the final round, carries market significance far beyond the deal itself. Bank of America Merrill Lynch points out that the price peak for software ETFs coincided precisely with NVIDIA's announcement of this investment in September 2023. NVIDIA's withdrawal now is a potential leading indicator that the exponential growth in AI capital expenditure may be decelerating. Once this trend is confirmed, it could become the optimal catalyst for reversing two major trades: the first is the "short tech bonds" trade (represented by widening Oracle CDS spreads); the second is the "long semiconductors, short software" trade (i.e., the "AI awe > AI poverty" logic). Bank of America Merrill Lynch emphasizes that a bottom in the software sector is crucial, as its performance is highly correlated with private credit and bank loans. This week, bank loan funds saw their largest outflow in three months ($900 million), and a bank loan ETF (BKLN) approached a "credit event" threshold. Strategists believe that the software ETF holding above $80 and the bank loan ETF holding above its February low of $20 are key technical supports for current market stability.
It is noteworthy that the Bank of America Merrill Lynch Bull & Bear Indicator remains at an extreme bullish reading of 9.2, signaling a sell. A global fund manager survey shows that emerging markets, European equities, and bank stocks remain severely overweight, meaning selling pressure on these assets should not be underestimated if the market declines further.
Comments