Summary

The speaker’s core judgment is that the tariff path is driven by political approval, not economic optimization. Trump pushed China’s tariff rate to 125%, while giving other countries a 90-day reprieve and temporarily cutting their rate to 10%, triggering a nearly 12% one-day rebound in the Nasdaq; but what he is really optimizing is his strongman image within the MAGA base. “How much stocks fall, and whether the economy gets a little better or worse,” are not decisive variables.
A 90-day pause cannot resolve long-term decoupling, and short-term implementation problems do not mean the policy will be withdrawn. Rules of origin, transshipment, WTO rules, foreign legislation and US Customs capacity all make a 125% tariff difficult to enforce; but once public opposition hardens, “there is no way to replace every share of trade that is lost” between the US and China.
Extrapolating the current policy “tangent” puts the probability of a US recession at roughly 50%, a risk markets have not fully priced. The speaker’s simplified math is that every 1 percentage-point increase in effective tariffs lifts inflation by roughly 0.1 percentage point and cuts growth by roughly 0.1 percentage point; current policy is enough to push inflation toward 3.5% and growth down to 1%. Semiconductors, electronics and pharmaceuticals could also face a 20%-25% critical goods tariff, while semiconductors and electronics may separately face a national security tariff.
If a recession materializes, a drop in the S&P 500 to around 4,000 would not be extreme. Goldman’s baseline puts this year’s EPS at $253 and 2026 EPS at $269, up 6%; in a stress period, forward P/E could compress to 13-15x. At 14x, “4,000 feels completely normal,” well below the market’s currently implied path of modest growth.
Trading discipline matters more than guessing the policy turn: keeping a small position is acceptable, but going all-in or buying rebounds off chart support should be avoided. The market is offering no visibility and delivering constant surprises; one or two negative surprises could wipe out leveraged positions. “No one pays for any of the stories in a bear market.” Cash should wait for genuinely cheap long-term valuations, rather than treating short-term support levels as a margin of safety.
On a relative basis, US mega-cap tech should outperform the S&P 500, which may in turn outperform China and Europe—but that does not mean tech is cheap. Tech companies have stronger cash flow, have already gone through one round of selling, and the US technology and corporate ecosystem cannot be replaced quickly. By contrast, the speaker believes China’s eventual share of exports destined for the US could be around 40%, close to 50%, versus roughly 20% in official statistics; and that portion represents manufacturing’s most profitable business, which ordinary consumer stimulus cannot replace.
The violent move in Treasuries looks more like large-scale deleveraging than a structural Treasury crisis. Five-day rate volatility was exceeded only in 2008 and 2019, while this year’s short-term issuance is around $1.7T, below $1.97T last year and $2.9T in 2023. Possible Chinese selling, Japanese hedge-fund selling, and the unwinding of basis and spread trades could create a spike of a dozen-plus basis points, but not permanently change supply and demand. The speaker believes recession risk remains underpriced in CDS, credit basis and equities.

Deep dive

1. Tariffs Are Optimizing for Political Loyalty, Not Economic Growth

The speaker shifts the market’s focus from economic data to Trump’s tariff policy: tariffs on China were first raised reciprocally to 84% and 125%, then reciprocal tariffs on other countries were temporarily cut to 10% with a 90-day reprieve, directly driving a nearly 12% rebound in the Nasdaq.
His political explanation is that Trump cares more about whether MAGA supporters see an “unyielding strongman” than about his overall approval rating in media such as CBS. The speaker calls CBS an extreme-left outlet and argues that Trump cares more about approval among Forbes readers, the far right and the MAGA base. “Politicians are optimizing for his approval rating, not the economy.”
That makes reports that Navarro is being set up as the scapegoat unconvincing. Bessent may have played some role in the rumors of tariff cuts, but the speaker does not believe Navarro will be pushed out of Trump’s inner circle as a result. He also notes that Navarro once “helped him do time in prison” and judges that Trump will not deal harshly with him.

2. The 90-Day Reprieve Has Only Moved the Battlefield; More Tariffs May Follow

Over the weekend, reports said semiconductors and electronics would be exempt, but Trump later denied it: the 20% rate belongs to the existing fentanyl tariff, China has been placed in a “different tariff basket,” and semiconductors and electronics may face an additional national security tariff. Future measures may target not only China but also places such as Taiwan.
Steel and aluminum tariffs are already in force. The next layer, a critical goods tariff, could cover semiconductors, electronics and pharmaceuticals at rates of 20%-25%. The 10% reciprocal tariff is therefore not the endpoint, but just one layer of the policy stack.
The bigger risk over the next 90 days is transshipment. If Europe, Japan, Mexico or Canada do not restrict Chinese goods, will the US block those channels? The US and other countries could then sanction one another, with policy repeatedly oscillating inside the 90-day window and continuing to generate risk.
But 125% is also extremely difficult to enforce in full. Identifying Chinese origin, tracking transshipment, breaking through the WTO system, persuading other countries to legislate and actually collecting the tariffs are all highly difficult. Even if the US can invoke emergency statutes, other countries’ presidents do not have comparable powers. A large number of short-term “bugs” does not mean the long-term direction has reversed.

3. The Current Policy Tangent Points to a 50-50 Recession Risk

The speaker admits he cannot predict Trump’s next move and can only draw a “tangent” to the policy curve. Assuming other countries remain at 10% and China’s tariff rate comes back down, he still assigns a roughly 50% probability to a US recession, above Goldman’s estimate of roughly 35%.
His adding math is that every 1 percentage-point tariff increase adds roughly 0.1 percentage point to inflation and subtracts roughly 0.1 percentage point from growth. Even before accounting for the self-fulfilling effects of a falling stock market, the current shock could push inflation to 3.5% and growth down to 1%.
With growth at just 1%, a single quarter of negative growth would be “completely normal.” This is the judgment under current policy; if all the policies are withdrawn, there will be no recession, and the assessment should be revisited once policy actually changes.
The long-term outlook is more bearish than the short-term one. Implementation details can be repaired, but once public sentiment in both countries turns confrontation into a source of political gain, “every share of trade that is lost between China and the US cannot be recovered.”

4. Equities Still Do Not Give Enough Weight to Recession and Valuation Compression

Goldman expects S&P 500 EPS of $253 this year, up 3%, and $269 in 2026, up 6%; the growth priced into the market is still around 8%-9%. The speaker believes that even with these relatively optimistic earnings forecasts, valuation compression alone could produce a major decline.
Looking back at stress periods such as 2018, 2020 and 2022, forward P/E typically fell to 13-15x, and to around 13x in particularly bad episodes. Applying a 14x multiple to 2026 EPS of $269 makes “4,000 completely normal” for the S&P 500.
A recession does not rule out a bull market beginning halfway through the downturn, but the most painful phase is often when recession expectations are forming and economic data are just starting to deteriorate. That is precisely the tail risk the market has not fully priced today.

5. In a Bear Market, Stories, Information Edges and Charts Offer Little Protection

Bull markets pay for new stories and informational advantages; information leaders in areas such as semiconductors can even continue generating excess returns. Bear markets work differently: “You think the story is great, you go in, and you get buried,” because the macro direction overwhelms the individual stock narrative.
The speaker’s warning is direct: “Don’t go in, don’t bet on the short term,” especially by buying a rebound based on chart support. Price-volume indicators are more useful when markets are relatively stable and information jumps are limited. Today’s market offers no visibility and keeps delivering surprises; one or two negative surprises could wipe out an entire position.
Existing positions that are modest in size do not need to be liquidated completely, because the S&P 500 may eventually recover; the holding experience and waiting time will simply be painful. The real danger is going all-in at the bottom. Valuation is “completely useless in the short term and always effective in the long term.” Cash should wait until valuations are cheap enough; if the market is pricing a recession, it is not cheap enough yet.
On relative positioning, he believes US tech companies have stronger cash flow, have already fallen once, and that the long-term technology ecosystem will not disappear quickly because of immigration or international-student policies. But “tech outperforming the S&P” is only a relative defensive conclusion; it does not mean either can no longer fall.

6. China and Europe May Not Be Safe Havens in a US Recession

The speaker’s valuation ranking is counterintuitive: Hong Kong first, followed by Europe and Japan, then the US S&P, with S&P tech last. His view is that the rest of the world’s valuation framework may ultimately suffer a more severe de-rating than the US.
Hong Kong is supported by Chinese consumption, while Europe is supported by European investment, but the US P/E framework cannot simply be applied to either market. The US valuation premium over the past 15 years came from the concentration of internet, AI and application companies—a growth structure that differs from China and Europe.
More important is the trade transmission mechanism. He estimates that the share of China’s exports whose ultimate destination is the US is not the roughly 20% shown in official statistics, but closer to 40%, approaching 50%. This is often the most profitable revenue for manufacturers and is used to subsidize other business lines.
Once that profit pillar is cut off, the shock will spread across entire companies and into employment. “Spending some money to stimulate consumption” or issuing more consumer credit cannot replace this external demand, so a US recession and 125% tariffs would drag down China’s fundamentals as well.

7. The Treasury Storm Is Deleveraging; The Real Endgame Requires Market Fatigue

Five-day Treasury yield volatility has reached levels seen in 2008 and 2019; only those two episodes were more extreme. On the fundamental side, tariffs are pushing up PCE and long-end inflation expectations. On the other, the market is worried that foreign investors will reduce Treasury allocations and weaken the dollar’s reserve status.
The technical factors may include Chinese funds selling Treasuries to repatriate capital and support Hong Kong equities, as well as several Japanese hedge funds liquidating Treasuries after losses in equities. The speaker believes it is probably not Norinchukin Bank itself, because a liquidity crisis there would affect the entire market far more severely than what we are seeing now.
Highly leveraged spread trades amplified the shock: positions that borrow short and buy long were forced to unwind, basis and repo rates rose, and tighter liquidity forced further long-bond selling. He has consistently advised against buying TLT; for a recession trade, he recommends CDS rather than simply betting on the long end.
Supply data do not support a structural crisis: short-term issuance this year is around $1.7T, below $1.97T last year and $2.9T in 2023; net-supply DV01 is roughly $758M per basis point, also below last year’s $1B. Even if China sells several hundred billion dollars, the main effect would be a short-term spike. The Fed can also cushion liquidity pressure through Treasury discounting or very low discount rates.
The 10-year yield peaked at roughly 4.4%-4.6%. If it now returns to 4.4%, the speaker sees that level as “borderline,” broadly fair value rather than an obvious crisis trade. The trade war will continue to “grind on”: deals, breakdowns, bug fixes and “great miracle day” rebounds will alternate, similar to the 2015 Chinese market deleveraging he cited. The process may only be nearing its end once everyone is exhausted, stops watching, and most of the necessary decline has already occurred.

Neil Mehta - Finding Future S&P 500 Companies - [Invest Like the Best, EP.419]110: Ming Ventures' Xia Ling on the Agent Gateway Battle | Agent #2