At one point this writer used to think that the mid-term elections were not so important to Donald Trump personally because he cannot run again for the presidency.
But it subsequently became self-evident that such an assumption was wrong when it was pointed out that the House of Representatives has the power to launch impeachments. For such a reason alone continuing Republican control of the House this November is a priority for the 47th president.
This is one reason, among many, why the decision to launch the attack on Iran on 28 February was so risky domestically as all the polling continues to show that such a policy was unpopular both that decisions and has continued to be after it.
The latest CNN poll conducted between 26-30 March found that 66% disapprove of the US decision to take military action in Iran, up from 59% in the previous poll conducted between 28 February and 1 March, while 68% oppose the US sending ground troops into Iran.
There is also the not unimportant point that Trump’s election campaign was all about ending foreign wars.
Then there is also the issue of Trump’s unfavourable polling as regards the economy in spite of economic data and a stock market which shows an American economy in comparatively good health even if a significant part of that growth remains driven by AI capex.
The Economist/YouGov poll conducted between 3-6 April shows that 57% of Americans disapprove of Trump’s handling of the economy compared with only a 39% approval rating.
While the CNN poll conducted between 26-30 March saw 69% of Americans disapprove and 31% approve of the president’s management of the economy .
AI is the One Bright Spot in the US Growth Story
As for AI capex, the latest 4Q25 US GDP data shows no sign of any pickup in capex outside of AI.
Real AI-related capex, measured as non-residential private fixed investment in information processing equipment, software and data center construction, contributed 1.11ppts or 56% to US real GDP growth of 2.0% YoY in the four quarters to 4Q25.
US real AI-related capex rose by an annualised 18.5% QoQ in 4Q25 and was up 19.6% YoY, while non-AI capex declined by an annualised 8.0% QoQ in 4Q25 and was down 3.0% YoY.
Similarly, real AI-related capex rose by 26.8% over the past two years to 4Q25, while real non-residential private fixed investment excluding such AI-related capex declined by 4.9% over the past two years.
Tariffs Unsurprisingly are Unpopular with the Average American
Meanwhile, it has become clear in recent months that Trump’s tariff policies were increasingly unpopular long before the Supreme Court decision on 20 February to declare most of them illegal.
Those are the “reciprocal tariffs” imposed under the so-called International Emergency Economic Powers Act (IEEPA).
Thus, an Economist/YouGov poll conducted between 15-17 November showed that 47% of respondents want tariffs on foreign goods to be reduced and only 13% want them to be increased.
While 73% of Americans said that Trump’s tariffs have increased the prices they have paid either a lot (40%) or slightly (33%).
It is also the case that the more recent Economist/YouGov poll conducted between 20-23 February, following the Supreme Court decision, showed that 60% of US adult citizens disapproved of the way Trump is handling tariffs, while 57% approved of the Supreme Court’s decision on tariffs with only 23% disapproving.
A total of 71% of Americans also said Trump’s tariffs increased the prices they paid for things.
This reaction is not so surprising since the longer the tariffs are in place, and they were then running at an annualised US$376bn based on October data, the more likely most of the bill will be paid for by the American consumer.
These political pressures explain why the Trump administration in November suddenly cut tariffs on items like beef, coffee and bananas to the benefit of a country like Brazil which, it should be recalled, was suddenly hit by 50% tariffs in late July because Trump did not like the treatment of former president Jair Bolsonaro.
Thus, Trump exempted in November food items such as beef, coffee and bananas from the “reciprocal” tariffs imposed on most US trading partners.
Meanwhile, monthly government receipts on tariffs have declined since the Supreme Court decision from a peak of US$31.354bn in October to US$22.155bn in March.
Tariff Refunds Could Spook the Treasury Bond Market
As regards to the Supreme Court, it might have been assumed that, since Trump appointed three of the nine Supreme Court justices with another three appointed by previous Republican presidents, that America’s highest court would not rule against Trump even though the legal argument of the administration is tenuous in the extreme.
This is because under the US Constitution, Congress oversees taxation, not the president.
The argument of Trump’s legal team on tariffs was that the US trade deficit, which has been a reality for 50 years, is a “national emergency” justifying the unilateral tariffs imposed by the administration.
For the record, the tariffs now average 11.8% based on the latest announced tariffs as of 8 April, the highest level since 1941, according to the Yale Budget Lab.
Still, the Supreme Court decision will now lead to a huge refund bill for the Treasury totaling about US$166bn, resulting in an administrative and bureaucratic nightmare, not to say a costly one. Total tariff collection was US$318bn in the 12 months to March, or a US$235bn increase from US$83bn in the previous 12 months.
An added complication is that the tariff revenue is now part of the US General Fund, which Congress controls, though it is managed by the Treasury.
Certainly, the sudden realisation that the tariff revenue may no longer be “money good” is also a potential negative for the Treasury bond market for those who view the increased revenues from tariffs as one explanation for the relative calm of the long-end of the Treasury bond market in recent months compared with some other G7 markets.
Meanwhile, another sign that Trump has begun to react to the negative impact of tariffs politically is that he proposed in November giving every American, excluding “high-income people”, a dividend of “at least” US$2,000, presumably to compensate for the price hikes caused by tariffs.
In addition to the precedent from Covid of what such handouts can do in terms of exacerbating inflationary pressures, such a policy would also be a negative in terms of the fiscal impact.
This is because such a handout is estimated to cost at least US$400bn. A study in November by the Yale Budget Lab estimated that a one-time US$2,000 per person rebate for Americans earning less than US$100,000 per year would cost US$450bn.
While the Supreme Court decision makes such a policy less likely, anything is possible in terms of stimulatory policies domestically in the run-up to the mid-terms, most particularly as the Iran War has already caused a noticeable pickup in inflation as a result of the surge in energy prices.
The latest US CPI data shows US headline CPI inflation increasing from 2.4% YoY in February to 3.3% YoY in March, and energy CPI inflation rising from 0.5% YoY in February to 12.5% YoY in March.
Meanwhile, it is interesting to note that the mid-terms are the year in the presidential cycle when share prices rally the least.
On that point, in the past 24 presidential cycles since 1929, the S&P500 rose by an average 3.3% in the second year of a cycle with only 54% of those years posting a positive return.
So far year to date the S&P500 is up 0.6%.
With near-term inflationary pressures rising in the US courtesy of the Iran War, the potential risk for the US stock market is a self-off in the Treasury bond market out of the recent tight trading range.
This has begun to happen.
The ten-year Treasury bond yield has broken out of the range which has been prevailing since last August, reaching a peak of 4.48% on 27 March and is now 4.28% (see following chart), as bond investors have focused anew on Iran-driven stagflation risks rather than AI-triggered white-collar job losses which was the main concern prior to the attack on Iran on 28 February.
Any move in the ten-year yield beyond the 4.5% level will surely be negative for equities, as well as causing investors to refocus on America’s fiscal issues.
It is also the case that Treasury bond market volatility has picked up of late, as reflected in the MOVE index. It surged by 28% on 20 March to a high of 108.84.
This was the biggest one-day gain since October 2020 during Covid.
The index rose to a recent high of 115.02 on 26 March, though it has since declined to 74.42.
Prior to this, the Treasury bond market has been benefiting from Treasury Secretary Scott Bessent’s ever-increasing resort to financing at the short end, as well as buybacks at the long end.
On buybacks, in the latest quarterly refunding review in early February, the US Treasury estimated that buybacks, including liquidity support and cash management buybacks, will increase from US$50bn in 1QCY26 to US$53bn in 2QCY26.
Meanwhile, gross Treasury bill issuance totaled US$26.27tn in the past 12 months to March, accounting for a massive 84.9% of the total gross marketable Treasury issuance of US$30.95tn.
Rising Yields Due to Iran Conflict are Making it Harder for Trump to Cut Debt Servicing Costs
The other point is that the energy price shock stemming from the Iran War means that financial markets are less convinced than they used to be that the Fed will be cutting rates this year.
Since the commencement of the Iran War, money markets have gone from assuming 61bp of Fed easing this year to 14bp of Fed tightening in late March and are now discounting only 9bp of Fed easing.
The key focus of the Trump administration remains on reducing debt-servicing costs, with net interest payments and entitlements rising to 92.7% of total federal government revenues in the 12 months to March, though down from a recent high of 95.6% in the 12 months to January 2025.
An added negative in this respect is that the flow of tariff revenue should be drying up as a result of the Supreme Court decision.
As already noted, monthly tariff revenues have declined from a peak of US$31.354bn in October to US$22.155bn in March, down from US$27.742bn in January prior to the Supreme Court decision.
The views expressed in Chris Wood’s column on Grizzle reflect Chris Wood’s personal opinion only, and they have not been reviewed or endorsed by Jefferies. The information in the column has not been reviewed or verified by Jefferies. None of Jefferies, its affiliates or employees, directors or officers shall have any liability whatsoever in connection with the content published on this website.
The opinions provided in this article are those of the author and do not constitute investment advice. Readers should assume that the author and/or employees of Grizzle hold positions in the company or companies mentioned in the article. For more information, please see our Content Disclaimer.