2026 Outlook: Duration Tightening

Run it Hot, Privatizing the Fed's Balance Sheet, Dollar Devaluation and the AI Boom

Key Themes, Forecasts & Risks:

• Bank deregulation and a steeper yield curve (3-month-10-year) release bank reserves and drive asset growth with significant macroeconomic and asset pricing effects.

• The FOMC reduces the policy rate to 3-3.25%, the impact on small banks and small businesses is greater than expected, both for the profitability and valuation of spread sensitive regional banks and the labor market.

• The FOMC also increases the pace of duration tightening by expanding reinvestment of mortgage paydowns into bills to a portion of maturing Treasuries.

• As a consequence of the Fed regulatory and monetary policy actions there is no need for additional balance sheet expansion, instead the privatization of the Fed’s balance sheet takes hold.

• Capital spending broadens to include non-AI infrastructure manufacturing.

• Consumption recovers as the effects of the three adverse aggregate demand shocks, tariffs, slower government spending and reduced immigration, fade and the individual tax provisions of One Triple B spur spending.

• Labor market demand and dynamism improve as Fed easing reopens the small bank credit channel and eases pressure on small floating rate borrowers. The increase in the U3 unemployment rate stalls at 4.75% in 1Q26 but doesn’t decline much due to structural pressure on employment (technology innovation adoption).

• Supply pressures return to the belly of the Treasury curve, 10s end the year at 4.5%, 2s at 3.4%, and with the policy rate at 3-3.25% the 3m10y curve crucial to regional banks ends the year above 1%.

• The privatization of the Fed’s balance sheet, bank deregulation, rate cuts and duration tightening gets into lots of market cracks. The yield curve steepens; cyclical stocks, metals, energy and industrials outperform.

• The trade weighted dollar has a similar ~8% decline led by Asian currencies as their trade surpluses stall and begin to contract. If the process stalls, a global accord is possible.

• The S&P 500 has a significant pullback in 1Q26 and ends the higher with about half the ‘25 gain.

• Credit spreads widen due to AI infrastructure debt. Fixed income supply from the Fed (DT), mortgage and credit markets, along with reduced global demand, cause a couple of real rate risk-off shocks.

Quick Review: Holding Ourselves Accountable

2025 Outlook: Targeting a Trifecta

Our key market forecasts for 2025, a 10% return for the S&P 500 led by cap-spending beneficiaries, 10-year USTs ending the year at 4.2%, 2s10s steepening to 70bp, the current coupon agency MBS rate at 5%, generally worked out well. More importantly, because our approach to investing is not ‘set it and forget it’, and year-end forecasts are only intended to quantify an outlook, we caught the only significant correction in late February and the market low in early April. Consequently, if you followed us, you should have done well. We avoided consumer sectors and leaned into investment sensitive sectors. The labor market stayed weak as we expected. The trade shock was larger than we thought was likely. The Fed and Treasury did not begin coordinated action to manage the pandemic debt blowout, but that process is beginning. Capex began recovering from the Biden Administration’s industrial policy mess, but only in AI infrastructure. Bank regulatory policy loosening took longer to get started than we expected but is likely to be a major theme in ‘26.

Run it Hot

One of the most widely expected themes for ‘26 is ‘they will run it hot’. They, in this case, refer to the Trump Administration and GOP controlled Congress. Setting aside our aversion to the assumption that ‘they’ have control over the economic outcome, as a base case we expect the effects of the three adverse aggregated demand shocks, lower immigration, slower government spending and higher tariffs, to fade. As the policies that were integral to the first second term president being elected with control over Congress since FDR morph into initiatives targeting the midterms, a modest recovery in consumer spending, stronger capex and increased labor market demand and churn are reasonably probable. Additionally, with the Fed on track to reduce the policy rate close to our estimate of neutral (a bit above 3%), the normalization of the upward sloping yield curve will reopen the small bank credit channel, and a recovery in small business employment and housing construction will contribute to stronger, more broadly distributed, growth in ‘26.

Figure 1: The upturn in S&P Small Cap earnings estimates is encouraging for the run it hot, stronger growth in ‘26 outlook. We did not use the Russell 2000 because the annual reconstitution causes significant changes to the index that makes comparisons to S&P indices awkward. The lack of an upturn in the Midcap suggests the upturn is far from the foregone conclusion you’d be led to believe listening to financial television.

While we expect the Fed’s yield curve normalization to close the profitability differential between large and small banks, and large and small companies generally, analysts do not expect the dispersion of earnings growth within the S&P 500 to change much. The technology and communication sectors are expected to be the primary drivers; energy is the largest drag followed by weak earnings from consumer sectors with solid, but unspectacular growth from financials and industrials. Healthcare earnings rebounded in ‘25, but the recovery is expected to stall in ‘26. A big wildcard for the earnings outlook is the trade weighted dollar; it is down 6.5% in ‘25 despite the increase in tariffs that boosted the dollar in ‘18 and relative monetary policy easing, hinting that a broader structural revaluation of the overvalued dollar is underway. We will discuss this in greater detail later in the note.

Figure 2: Fundamental analysts, due to cautious guidance from companies, are not as bullish on the ‘healthy broadening out’ as strategists. One year-forward estimates are increasing at a point when they’d typically be falling though.

Our base case sounds very much like a decent year for equities; however, as has been the case frequently since the pandemic fiscal and monetary policy panic increased federal debt service to the second largest budget item, the path between weaker growth and restrictive real rates is narrow. We’ve heard a lot about the Fed restarting QE in ‘26, particularly from the crypto crowd, as reason for financial asset bullishness alongside the ‘run it hot’ scenario, but our outlook for monetary policy is for less accommodative policy for long duration assets.

Policy Outlook:

“The Fed is independent within, not of, the government” William McChesney Martin, Chairman of the Federal Reserve Bank 1951-1970

As we were writing the note ADP reported a 120,000 drop in small business employment, pushing the annualized rate into a recessionary 0.24% contraction. A theme in our work since the Fed belatedly responded to the ‘21-’22 fiscal inflation shock they facilitated with their reckless purchases of Treasury notes and mortgage-backed securities was the pressure on small banks and businesses from the Fed’s clumsy tightening process. Their misguided tightening tactics, passive balance sheet contraction that impacted the least efficacious asset purchase channel, bank reserves, combined with the most aggressive rate hike cycle since the Volcker Fed, tightened conditions sharply for floating rate borrowers, while leaving policy accommodative for fixed rate borrowers. The consequences were profound, a ‘K’ shaped business sector unfolded with small banks, businesses and leveraged borrowers bearing the brunt of tighter rate policy, while asset rich households, high quality fixed rate borrowers and large banks benefiting from the Fed’s balance sheet continuing to put downward pressure on longer term rates.

Figure 3: Duration of the high yield index began shortening when the Fed started hiking, and the spread relative to investment grade credit is near record levels. This illustrates the ability for high quality to term out their debt, while more leveraged companies struggle with higher short-term rates. The MBS index duration is near record levels underscoring how asset rich households also benefited from the Fed’s unbalanced tightening cycle. This is the K-shaped economy.

Our market derived estimate of the neutral policy is marginally above 3%. When the Fed hiked above 3% in September ‘22, exchange rate and fixed income market volatility exploded forcing the Fed to slow the tightening process. Additionally, the Fed’s portfolio began losing money and losses in the banking system’s securities holdings led to the collapse of Silicon Valley Bank and forced sale of First Republic to JP Morgan. Over the last four years when it became obvious the Fed had made a massive policy mistake in ‘21 we have not heard a single member of the FOMC mention the shape of the yield curve and impact on the banking system. This is basic banking, to justify taking on credit risk, even sovereign risk, requires a term premium for holding longer-term maturity assets.

Figure 4: The median of the 3m10y banking proxy curve monthly average since 1941 is 1.41%. The median of the 10-year note yield is 4.26%. Consequently, to illustrate how exceptional and restrictive the current policy is, the current 3m10y curve at 0.24% is 1.03 standard deviations below its historical median, however the current level of 10s is only 0.07 standard deviations below its median. The combination of a flat curve for the basic deposit loan banking model and low longer yields is attributable to the Fed’s large holdings of longer maturity Treasuries and mortgage-backed securities and restrictive rate policy.

Monetary Policy: Privatizing the Balance Sheet

“Overall, most participants favored a long-run composition of the SOMA portfolio that matched the composition of Treasury securities outstanding, indicating that a proportional allocation would provide enough flexibility and may be simpler to communicate. Some participants indicated that they favored a larger-than-proportional share of Treasury bills, citing the benefits of having even greater flexibility than available under a proportional allocation”. Minutes of the FOMC Meeting October 28–29, 2025

Like the majority of FOMC, market participants generally focus on rate policy rather than the balance sheet. Asset purchases impact markets through three channels, bank reserves (QE), duration (DE) and volatility suppression (VE, though this is the first time we called it that) through the purchases of mortgage-backed securities. The Fed’s tightening process ignored the most powerful channel, duration as well as volatility that remains abnormally low. We learned in the early ‘00s when the Bank of Japan injected 30 trillion of reserves into the Japanese banking system, even after a government recapitalization of the sector still struggling with the bursting of the twin bubbles more than a decade earlier, that QE had virtually no impact on the supply of credit and economic activity. We do not believe it was a coincidence that on the first day of duration tightening (DT), December 1, when the Fed began reinvesting mortgage paydowns (~$17 billion per month) into Treasury bills, rather than notes or mortgage-backed securities, the yield curve bear steepened. The statement from the minutes of the October FOMC meeting suggest the Committee plans to reinvest maturity Treasuries into bills to match the term structure of the outstanding stock of Treasuries. Suppressing duration and volatility is a one-way ticket to capital misallocation, or malinvestment as the Austrian School refers to it. An example is private equity investment in residential real estate and the affordability crisis.

Figure 5: Treasury Secretary Yellen’s increased bill issuance isn’t historically unusual, if we were in recession. The administration expects the growth of stable coins following legislation to absorb a portion of the bill issuance.

The System Open Market Account holds $6.13 trillion of securities outright, $3.56 trillion of Treasury notes and bonds, $2.045 trillion of MBS and only $195 billion of bills. To increase bills to 32.7% of their holdings of USTs requires reinvesting $1.03 trillion of maturing notes into bills, to get bills to the same percentage of the entire SOMA portfolio increases the amount of bill reinvestment to $1.87 trillion. Were the FOMC to rebalance the Treasury portfolio in one year implies $86 billion per month, far more aggressive than is likely. The FOMC will no doubt move slowly, however we ran through these numbers to illustrate the magnitude of the problem they created with their reckless pandemic purchases. Something like $25 billion per month of maturing coupons reinvested into bills seems like a good place to start.

Figure 6: Small banks haven’t bought securities, or grown assets, meaningfully since the regional bank crisis in early ‘23. A steeper curve is likely to improve demand for securities and increase loan growth.

In October 2023, following a 100bp increase in 10-year Treasury yields, after the last rate hike in the cycle, Treasury Secretary Yellen reduced issuance of notes and bonds, and increased bill issuance above 30% despite the Treasury Borrowing Advisory Committee’s earlier 15-20% guidance. The tactical issuance change stopped the supply driven selloff, but the Biden Administration and Congress did not take any steps to slow the root cause, government spending as percentage of GDP 4-5% above its 50-year median. Fed Governor Miran wrote a paper calling the issuance changes activist treasury issuance (ATI).

Figure 7: Banks bought Treasuries aggressively right alongside the Fed even at record low yields. The combination of excess reserves and an upward sloping curve was a powerful incentive. Bank deregulation should create excess reserves and rate cuts should steepen the curve. The Treasury General Account is a sideshow that investors, particularly the crypto crowd, should ignore.

There are two policy steps planned by the Trump Administration to find a home for the Fed’s holdings of longer duration securities and perhaps allowing Treasury to reduce their reliance on bill issuance as well. The first, a steeper yield curve, requires cooperation from the Federal Reserve. While many investors, economists, FOMC participants, and Fed staffers think a lower policy rate risks higher inflation, we disagree, particularly if it is combined with shortening the duration of the Fed’s portfolio. The second step is banking sector regulatory policy loosening. In Congressional testimony last week, Supervision and Regulation, Vice Chair for Bank Supervision Bowman detailed the four major areas of focus, stress tests, the supplementary leverage ratio, Basel III and the G-SIB surcharge. We have seen numerous estimates that bank regulatory policy reform could facilitate ~$6 trillion in balance sheet expansion. The 2017 Mnuchin Treasury bank reforms released ~$4 trillion of bank balance sheet capacity, 85% of which did not require legislation.

Figure 8: November was the second biggest month for IG credit issuance, assuming our data from Bloomberg is correct. Apologies for the date format, a summer analyst built this spreadsheet, and it worked great until this week.

The process of privatizing the Fed’s balance sheet, and reliance of the Treasury on bill issuance (unwinding Yellen’s ATI), can probably be managed without significant market disruption by the Bessent Treasury. Consequently, this process is unlikely to devolve into a disorderly bear steepener, like the ‘23 episode. What they have less control over is private sector supply of longer duration debt. November marked a turning point in AI infrastructure financing when investment grade issuance increased from $150 billion in October, a trend level of supply, to $210 billion. The increased supply did not widen spreads across the entire credit universe, and capex as a percent of cash flow in the technology and communication sectors remains far below the 2000 peak. Nevertheless, we expect increased AI infrastructure debt, mergers & acquisitions financing, an increase in manufacturing capex, and increased mortgage supply. The supply of duration is a significant risk to equity and fixed income investors in the ‘run it hot’ scenario. One street estimate we saw estimates $3 trillion of AI related capex by 2030, $1.4 trillion from hyperscaler cash flow, the balance from credit markets.

Summary: We expect another 75bp reduction in the policy rate in ‘26 to a 3.0%-3.25% corridor and an increase in DT to include $25 billion per month in Treasury bill reinvestment of maturing Treasury coupons.

Market Implications: A steeper curve, weaker dollar relative to Asian exporters, the euro and pound, and higher commodity prices, and higher Treasury yields in the belly of the curve as ownership transfers from a price insensitive buyer (the Fed) to price sensitive buyers (banks).

Fiscal Policy: Selling the Three Adverse Aggregate Demand Shocks

A year after President Trump was elected with control of Congress in a second term for the first time since FDR. The three adverse aggregate demand shocks, immigration, trade and government spending, were critical economic election issues in the President’s re-election, however, the near-term costs slowed growth, reduced employment and have lowered the probability of the GOP maintaining control over the House of Representatives. The fiscal impulse will turn more favorable when the individual provisions of the One Triple B Act impact consumers through refunds beginning in March.

Figure 9: Sharp drops in confidence matter as recession signals and are decent election indicators. This suggests the Speaker Johnson is likely to be the minority leader at this time next year.

“The inflation response goes against the predictions of standard models, whereby CPI inflation should go up in response to higher tariffs. Instead, tariff shocks appear to act as aggregate demand shocks—moving inflation and unemployment in the same directions. A possible explanation relies on the effects of uncertainty: a tariff shock creates (or coincides with) an uncertain economic environment, which by itself depresses economic activity by lowering consumers’ and investors’ confidence and puts downward pressures on inflation (Leduc and Liu, 2016). Another possible channel is a wealth channel, whereby an adverse tariff shock leads to a drop in asset prices, which then depresses aggregate demand and leads to higher unemployment and lower inflation. We find evidence in support of both channels: in response to higher tariffs, stock prices decline and stock market volatility increases”.

What Is a Tariff Shock? Insights from 150 years of Tariff Policy

We are in strong agreement with the conclusions of the San Francisco Fed staff’s paper. As of this writing we are waiting for the Supreme Court decision on the IEEPA tariffs, our core assumption is the current effective tariff rate will persist, though we do expect a decision that limits the Administration’s degrees of freedom. We were struck by a comment from Commerce Secretary Lutnick that their analysis concluded tariffs less than 15% had little impact on price. We took this to imply they were unlikely to raise the aggregate effective tariff rate further. Consequently, as we’ve been suggesting for months, adverse aggregate demand shock, has been absorbed. Whether import substitution and onshoring will reduce the trade deficit will become evident over the rest of President' Trump’s term.

Figure 10: When the post-tariff trade data finally gets tabulated, we expect to see demand for foreign goods flatline. we suspect we are close to the effective tariff rate peak.

If demand for labor continues to deteriorate, leading to an unemployment rate above 4.75%, consumption stimulus along the lines of President Trump’s $2000 tariff rebate checks, is increasingly probable. This policy step would be a major setback in one of Treasury Secretary Bessent’s ‘Three Arrows’, it would increase government spending and would increase inflation in our view.

Figure 11: We will be watching in ‘26 to see if the One Triple B eligibility requirements slow Medicaid spending growth.

While the bottom-line 3% spending growth for fiscal year ‘25 was sharply lower from 11% in ‘24, and well below the longer-run median rate of 6.5%, the chart above and table below shows the most inflationary components, direct transfer payments to individuals continue to expand well above the growth rate of the aggregate economic demand or income. There are provisions in One Triple B, work requirements for Medicaid most notably, that may slow growth over time. There is no change to our core view that federal debt stabilization can only be accomplished by lowering outlays to the 1981-2019 median at 20.3% of GDP. To achieve that objective, the growth rate of spending needs to remain below nominal output, and productivity needs to increase the real rate of growth. Both are possible, but not yet probable.

Figure 12: The declines in spending look more like one-offs then real reform.

From a ‘25 perspective, the drag from the three adverse aggregate demand shocks should fade, setting the stage for stronger disinflationary growth. Whether these major policy changes will achieve their longer-term objectives including facilitating a manufacturing renaissance will be far from definitive, but the signs from capital spending should be encouraging.

Economic Outlook

In February we wrote, Cleaning Up the Industrial Policy Mess, our thesis was that the process Secretary Bessent called detoxing from our addiction to government spending, would weaken growth and employment until 4Q. The magnitude of the trade shock and government shutdown may have delayed the recovery, but we still expect a more investor friendly, disinflationary, mix of growth to result from stronger capex and lower government spending in ‘26. In short, a more productive economy.

Figure 13: This is an exceptionally stale chart; however, gross output is the broadest measure of economic activity and for private goods producing industries value-added nominal output fell into contraction in 1H26. There may have been a tariff front-running bounce early in 3Q, for all of ‘25 it’ll probably be near zero growth.

Capex: Not Just AI

The Biden Administration and 117th Congress industrial policy legislation, dubbed modern supply side economics by Treasury Secretary Yellen, led to a sharp downturn in private sector capital investment in the major contributors to nonresidential fixed investment. A contributing catalyst for the downturn in equipment and R&D investment was the decision to allow the Tax Cuts & Jobs Act (TCJA) tax incentives to expire. There appears to be a recovery underway, however it is highly concentrated in the AI infrastructure boom and the improvement in 1H25 is flattered by tariff front-running.

Figure 14: The rebound in capex in 1H25 was highly concentrated and flattered by front running tariffs. Structures investment, despite the data center boom, is very weak.

Before moving on to the ‘26 outlook it is important to consider longer run trends. Structures investment was consistently between 3 and 4% of GDP from the end of WWII until the ‘90s. The ascendency of China and to a lesser extent the former Soviet Bloc led to outsourcing and a lower trend rate of physical plant investment. Investment in technology was an expense item in the national accounts data (GDP) until 2013, at that point the BEA revised the series and categorized software and R&D as investment. Equipment investment is at the lower end of the longer run range, underscoring that the lengthy ‘10s expansion was the second weakest capex cycle since WWII. In short, there is plenty of scope for a secular increase in capital investment across all categories.

Figure 15: This chart of investment relative to GDP shows the impact of offshoring.

Capital spending plans, as measured by regional Federal Reserve Bank manufacturing 6-months forward capital spending surveys, surged following the passage of the TCJA. Two factors stalled the advance. First, the investment incentives were a second order effect of TCJA, the primary impact was the increase in cash flow resulting from the lower corporate tax rate that led to a surge in stock buybacks (returned to shareholders). This was partially attributable to the technology sector that did not benefit from the lower tax rate because their use of transfer pricing tax arbitrage meant their effective tax rate was already at the new statutory rate (21%). They received a one-time windfall from the reduced tax rate on repatriating the profits held overseas, they largely used the cash for stock repurchases or to paydown debt that was an element of the tax arbitrage. The second factor that stalled a recovery in capex was the US/China trade war, capital spending plans fell after each of the major Trump Administration tariff hikes.

Figure 16: The primary impact of TCJA was lower taxes boosting buybacks. One Triple B did not lower the corporate tax rate, but it made immediate expensing of equipment and R&D permanent. The temporary manufacturing provision should spur structures investment.

The three aggregate demand shocks are the most probable explanation for muted capex plans since the July 4 signing of One Triple B. That said, over the last couple of months the regional Fed manufacturing surveys are gaining momentum. While service providing industries are lagging, the improvement in manufacturing orders and the employment outlook are encouraging.

Figure 17: Manufacturing capex plans are improving, but the services sector is struggling with weak final demand due to three adverse aggregate demand shocks.

Productivity

The ‘10s, not coincidently, were also one of the weakest productivity post-war productivity cycles, until the last two years. Stronger capital investment is critical to the outlook for productivity and inflation. The trade cycle has turned; we are in a period like the 1920’s, we are not going back to a 2.5% effective tariff rate. AI is likely to positively contribute to productivity growth over time. Additionally, we expect diffusion of technology innovation adoption to a broader range of sectors including the financial, healthcare and manufacturing sectors, a trend that was evident in consumer services in ‘18 and ‘19. The BEA GDP and GDI ‘25 data is a mess due to inventory and net export swings resulting from the trade shock, consequently any conclusions about productivity are speculative. What we do know is S&P 500 3Q25 annualized employment growth was -0.8% and sales per employee, outright and deflated using PPI, are surging. We suspect large company productivity is strong, small company productivity is probably not.

Figure 18: The sectors driving sales per employee, communication services, technology and financials, are the same early adopters of the internet in the late ‘90s.

Consumption

Spending on services is ~2/3 of personal consumption expenditures and they growth in spending slowed from 3.4% in September ‘24 to 2% a year later. This is the cleanest measure of demand; it removes any tariff price effect and reflects the adverse aggregate demand shocks.

Figure 19: Seasonal adjustment factors are corrupted by the pandemic, using the year-on-year rate smooths the data. Since Liberation Day the annualized rate of core retail sales has cooled from 5.5% to 3.3%.

The fading of the impact of the three shocks, trade, immigration and government spending, should give way to secular tailwinds. Despite all the concerns about the large swing in population growth and the aging population, life cycle consumption patterns are favorable for consumer spending. When the Boomers, the second largest age cohort, were in peak savings years from the late ‘90s through the mid ‘10s, and the smallest cohort, Xers, were in peak spending mode, trend consumption slowed. Currently, the peaks savers are the Xers, peak spenders are the largest cohort, Millennials, and Boomer’s marginal propensity to spend is increasing (can’t take it with you).

Figure 20: This is underappreciated in our view. Assuming linearity as the population ages is an oversimplification.

Labor Demand & Churn

As we began putting pen to paper (metaphorically) on our outlook note, our longstanding view that the FOMC’s misguided, unbalanced policy tightening, received an exclamation point in the form of a 120,000 monthly drop in small business employment in the ADP report.

Figure 21: Small business employment has been weakening for three years, and the rate of decline is accelerating, but the November 120,000 decline was shocking.

The BLS has been overestimating small business employment for years as evidenced by three consecutive negative annual benchmark revisions. We suspect the overestimation of 79,000 jobs per month for the year ended March ‘25 got larger in the aftermath of the trade shock.

Figure 22: The BLS has lots of explaining to do, they have finally taken a step to sort their overestimation of small business employment out.

With the market and FOMC coming around to our view that labor demand is falling faster than the immigration policy reduction in supply, this is the single most important indicator for the ‘run it hot’ outlook. We expect some additional upward pressure on unemployment, perhaps to 4.75%, before demand and churn (dynamism sounds better) improves in ‘26. The good news is if the primary catalyst for weakness is the Fed’s unbalanced policy, with the Trump Administration’s three shocks exacerbating the pressure on small businesses. The policy driven weakness is good news in the sense that additional rate cuts and the fading effect of the three shocks should lead to a recovery in the small business sector in ‘26. We absolutely reject the notion that another rate cut or two don’t matter. An upward sloping yield curve is crucial for the small bank credit channel and residential real estate market.

Figure 23: We are anxious to see whether this is false dawn, like the early year stabilization in churn, or demand for low wage workers and churn are stabilizing.

Not only did the Fed’s unbalanced policy weaken demand, it crushed churn, as measured by the spread between the wages of job switchers and stayers, and our favorite measure, worker reallocation (the quarterly sum of hiring and separations expressed as a percent of the labor force). We haven’t seen the JOLTS report since August so we will set that indicator aside for now, but the switchers-stayers spread in the Atlanta Fed Wage Tracker series began recovering this summer.

A technical measurement change could put downward pressure on monthly payrolls in early ‘26 but should also capture a recovery on a timelier basis. The BLS announced that beginning in January they are going to add monthly sampling to the birth/death model estimate of small business employment.

Disinflation

Lower trend inflation is probable through 1H26 primarily due to the lag impaired rent of shelter measures in CPI and PCED. The November ‘24 through January ‘25 comps (0.3%, 0.4% and 0.5%) are likely to weaken the how can the Fed cut rates with inflation at 3% narrative. More importantly the increase in core goods prices is likely to stall, and the sector most sensitive to fiscal policy, non-housing services, is unlikely to respond to the positive fiscal impulse until 2H26.

Figure 24: The trends in inflation were more concerning when the Fed eased in ‘24. We thought the cuts were appropriate, the increase in rates in the belly of the curve had nothing to do with the Fed as evidenced by breakeven inflation rates. The Treasury market correction was driven by election related supply concerns.

The private measures of rents continued to deteriorate in October and November, for example the Apartment List New Rent Index was -1.2% in November with vacancies at a multi-year high. The CPI rent of shelter measure will likely contract below 2.5% in 1H26, if it doesn’t the BLS’s new management should fix that also.

Figure 25: Here is another project for the BLS.

Disinflation in the government inflation measures should be a tailwind for monetary policy and allow the privatization of the Fed’s balance sheet to proceed in 1H26. We do not expect inflation to return to 2%, as we’ve mentioned numerous times, during the last 3 decades there was only one period when it was below the Fed’s target. The catalyst for the target was the ill-advised deflation fight in the aftermath of the financial crisis. Prices are stable, and the current rate is right at the post-war median rate of 3%.

Summary: Stronger capital investment, faster productivity growth, slightly stronger consumption, increased residential investment, a later recovery in the labor market and disinflation, are favorable for earnings. If the recovery does not materialize early in ‘26, additional fiscal stimulus could be a major setback on the path to reducing government spending to 20% of GDP.

Asset Allocation

Bonds: Bear Steepening

Duration tightening, alongside capex driven increased demand for credit, is likely to be a significant theme for markets in ‘26. We’ve heard investors and FOMC participants dismissive of the impact of the duration of the Fed’s portfolio. The arguments generally can be simplified to it hasn’t mattered, so why will it now? That argument is weak inasmuch as the FOMC, with the exception of allowing their mortgage holdings to shrink by $200 billion in 2010-2011 into a rally in Treasuries, has never reversed duration accommodation with symmetrical tightening. In other words, they did a lot of buying but never sold any of the securities, they merely allowed them to mature or in the case of mortgages get paid down, mostly because rates declined. The best example of why duration matters is the 120bp increase in 10-year yields from August through October 2023 when Treasury Secretary Yellen announced an additional, unexpected $500 billion of note and bond issuance.

The 2s10s term premium was deeply negative prior to the Fed beginning QT, the compelling explanation for this theoretical impossibility is the Fed’s price insensitive purchases. The term premium remains 50% below its long-term median. Assuming the Fed reducing duration is a non-event is reckless. With changes to personal at the Board, and maybe even the New York Fed (just a suggestion), the process could be handled adeptly and the longer run benefit of price sensitive buyers acting as check on profligate government spending could be profound. That said, while the Fed and Treasury can manage this process, demand for credit from AI infrastructure, capex broadening to include financing of structures driven by One Triple B’s manufacturing plant accelerated depreciation, and increased demand for credit from real estate developers. The offset is banking regulatory policy loosening, but the banking system may not open the spigot as quickly as policymakers would like.

Another factor that argues for a steeper curve is global sovereign debt markets. The Japanese are about to raise the policy rate, and though the hike is priced into the front end of the curve, the trend in longer maturity JGB yields is higher. Asian exporters have been put on notice, do not facilitate weaker exchange rates to avoid tariffs. Although talk of a Mar-a-Lago Accord has faded, if the trade deficit doesn’t contract next year it may resume. Demand for US fixed income is likely to slip in ‘26 from Asia. In Europe, Germany’s fiscal expansion is poorly designed in our view, the UK and France are struggling with their fiscal position, consequently those sovereign debt markets are unlikely to spark a global sovereign debt rally either.

We are convinced the yield curve is going to steepen again this year, and in our run it hot base case, bear steepening with 10s rising to ~4.5% is likely to be the catalyst for a decent sized equity market pullback in 1Q26, but not as deep as the 22% plunge following Liberation Day. We’ve raised a bit of cash, after we finally get the BLS and BEA data releases we expect to reduce risk further.

In ‘24 Treasury yields increased despite Fed policy rate easing due to supply concerns. As supply expectations contracted in ‘25 due to deceleration in government outlays, yields fell. With the Fed policy path discounted, unless the economy fails to respond to the One Triple B tax expenditures and easier rate policy, supply will likely put upward pressure on rates in ‘26.

Summary: The yield curve bear steepens in 1H26, as bank deregulation unfolds the belly of the curve stabilizes and ends the year with 2s at 3.25% and 10s at 4.25%.

Figure 26: Term premium is a model with plenty of critics; it is supposed to represent the premium for taking maturity risk. It could have never turned negative without the massive price insensitive buying from the Fed.

Stocks: Sitting Tight, For Now

The divergence between strategists expecting broader growth across sectors and capitalization and fundamental analyst estimates is likely to begin the year with strategist frustration. Consensus trades often get rinsed early in the calendar year, in ‘26 small caps, consumer discretionary and healthcare are our leading candidates for a shake out. We expect to increase our exposure to the run it hot theme at some point early in the year, perhaps following a disappointing 4Q25 earnings season combined with tepid management guidance.

Energy and materials are sectors we expect to have better years than in ‘25 due to a broadening of capex, however given that we expect the first response to improving growth to be bear steepening in the Treasury market we will wait for a better entry point.

Financials are our largest overweight, and while it might take time for looser regulatory policy to accelerate asset growth and return on equity, a steeper curve will have a more immediate effect, particularly on regionals.

Figure 27: The spread between ROE of large and small banks is a function of the shape of the curve, return on assets is virtually identical. Large banks are far less reliant on borrowing short and lending long.

We are not going to make a sector or risk changes simply because we are writing an outlook note. Our primary themes in ‘25 were avoiding consumer sectors due to the impact of tariffs on margins and leaning into investment sensitive sectors. The first effect should fade and of course the biggest issue for equity investors in ‘26 is managing exposure to the AI infrastructure theme. Our framework for determining what inning or year in the late ‘90s we are in is twofold. First, it is capex as the percent of cash flow in the communication and services sectors. Second, is monitoring the impact of AI infrastructure debt financing on credit markets. If and when spreads widen due to increased supply, that will be a sign the trend is extended. Meanwhile, the largest tech stocks were capital light, margin expansion stories since the financial crisis. Spending on data centers threatens to turn them into auto OEMs over time. We are underweight relative to our benchmark, but very exposed to tech and comm services in absolute terms. We won’t be swayed by anecdotes, for now we think we have a decent framework for assessing the stage of the cycle.

Figure 28: Industrials and communication services capex as a percent of cash flow tells the AI boom story.

Summary: The S&P scratches out a 5-7% return with the equal-weighted index outperforming (a call we got wrong this year).

Exchange Rates & Exceptionalism

One of the most surprising reactions to the Trump Trade Shock was the weakness of the dollar index and trade weighted dollar. The drop following Liberation Day ran in the opposite direction of the 10% increase relative to CNH (offshore yuan) following the 10% Chinese tariffs in ‘18. We suspect, based in part on the sharp move higher in the Taiwan dollar in early July, that exchange rates were integral components of the trade negotiations. For stability reasons, neither party will admit to discussions over exchange rates, however, the breakdown of the relative rate differential driver of exchange rates, notably the euro, suggests a broader correction in the structurally overvalued dollar (REER or PPP valuation metrics) is underway. Were the dollar to resume the decline that began after the trade shock, equity investors would initially react positively due to the impact on offshore earnings, however if the decline gathers momentum, the Treasury market struggles or foreign investors reduce their holdings of US equities, a weaker dollar could be the catalyst for a risk-off episode.

Figure 29: The drop in the dollar was unexpected following the tariff shock. We suspect a structural unwind of the dollar overvaluation is underway.

The structural overvaluation relative to Asia warrants a further decline, but appreciation of these exchange rates also suggests reduced demand for US fixed income. It isn’t as clear the euro should appreciate much. European economic dynamism is nowhere to be found at present, in part because their fiscal accounts are crippling and regulatory regimes are overly restrictive.

This brings us to the US exceptionalism theme, a term that hinges on investors outlook on US technology. The gap between return on equity for the S&P 500 and the rest of the world has been widening since the financial sector deleveraging ended in 2013/14. The US technology sector has been the driver. As the below chart shows there is no positive momentum in the rest of the world, with the possible exception of emerging markets. We have some small positions outside of the US, Chinese tech, Japan and Germany, but see now reason to increase exposure at this time.

Figure 30: This explains why US equity valuations are richer than the rest of the world, the companies make more money.

Final Thoughts

This was an exceptionally lengthy note, thanks for your patience in making it all the way to the end. We have two more year-end notes to write. Next up is our macro themes note, followed by our year-in-review note we began writing at Barclays to help clients with their year-end client communiques. We will release those on December 17 and 24, we may have some short updates in between as the BLS and BEA releases delayed economic and inflation reports.

Sector and Asset Allocation Tables Explained:

The US Equity Market Allocation table is our recommendations for a US equity investor, a similar approach to when we were the Head of Barclays US Equity Portfolio Strategy. The first six columns are valuation metrics, the seven is a Z-Score summary of the metrics relative to each sector’s valuation range since S&P introduced each sector (1990 for all but Real Estate). A reading of 1 implies the sector is 1 standard deviation above its historical median. The equity risk premium (ERP) column, also known as the Fed Model, is the forward (expected) earnings yield less the real 10-year yield (TIPS). Index weights are the S&P 500 with the exception of the Russell 200 small cap index, that is based on the market cap of the Russell 2000 relative to the Russell 3000. The final three columns are the Ironsides recommended weight, the 30-day volatility of the sector and portfolio contribution of our recommended weights to the risk (volatility) of the portfolio. Importantly, this approach does not integrate cross correlation of the sectors.

The asset allocation table benchmark is a 60/40 (stocks/bonds) portfolio, under the assumption that the investor is investing US dollars. We begin with our recommended weights, add the yield, the third columns are valuation metrics. ERP is the equity risk premium. TP is the term premium for Treasuries using the Adrian Crump & Moench model from Bloomberg. OAS is the option adjusted spread (early call risk) for fixed income securities. ‘SD’s’ from the median is a Z-Score approach to the valuation metrics, positive readings imply the asset is expensive, negative readings imply the asset’s valuation is below its longer-run median. The sixth column is the assets contribution to the risk of the portfolio, its volatility multiplied by the recommended weight of the asset. The index weights for equities use the same approach as the equity only portfolio, the fixed income weights are based on the Bloomberg US Aggregate Index, adjusted for the 60/40 benchmark. The final two columns are self-explanatory.

Barry C. Knapp

Managing Partner

Director of Research

Ironsides Macroeconomics LLC

908-821-7584

bcknapp@ironsidesmacro.com

https://www.linkedin.com/in/barry-c-knapp/

Follow us on X @barryknapp

This institutional communication has been prepared by Ironsides Macroeconomics LLC (“Ironsides Macroeconomics”) for your informational purposes only. This material is for illustration and discussion purposes only and are not intended to be, nor should they be construed as financial, legal, tax or investment advice and do not constitute an opinion or recommendation by Ironsides Macroeconomics. You should consult appropriate advisors concerning such matters. This material presents information through the date indicated, is only a guide to the author’s current expectations and is subject to revision by the author, though the author is under no obligation to do so. This material may contain commentary on: broad-based indices; economic, political, or market conditions; particular types of securities; and/or technical analysis concerning the demand and supply for a sector, index or industry based on trading volume and price. The views expressed herein are solely those of the author. This material should not be construed as a recommendation, or advice or an offer or solicitation with respect to the purchase or sale of any investment. The information in this report is not intended to provide a basis on which you could make an investment decision on any particular security or its issuer. This material is for sophisticated investors only. This document is intended for the recipient only and is not for distribution to anyone else or to the general public.

Certain information has been provided by and/or is based on third party sources and, although such information is believed to be reliable, no representation is made is made with respect to the accuracy, completeness or timeliness of such information. This information may be subject to change without notice. Ironsides Macroeconomics undertakes no obligation to maintain or update this material based on subsequent information and events or to provide you with any additional or supplemental information or any update to or correction of the information contained herein. Ironsides Macroeconomics, its officers, employees, affiliates and partners shall not be liable to any person in any way whatsoever for any losses, costs, or claims for your reliance on this material. Nothing herein is, or shall be relied on as, a promise or representation as to future performance. PAST PERFORMANCE IS NOT INDICATIVE OF FUTURE RESULTS.

Opinions expressed in this material may differ or be contrary to opinions expressed, or actions taken, by Ironsides Macroeconomics or its affiliates, or their respective officers, directors, or employees. In addition, any opinions and assumptions expressed herein are made as of the date of this communication and are subject to change and/or withdrawal without notice. Ironsides Macroeconomics or its affiliates may have positions in financial instruments mentioned, may have acquired such positions at prices no longer available, and may have interests different from or adverse to your interests or inconsistent with the advice herein. Ironsides Macroeconomics or its affiliates may advise issuers of financial instruments mentioned. No liability is accepted by Ironsides Macroeconomics, its officers, employees, affiliates or partners for any losses that may arise from any use of the information contained herein.

Any financial instruments mentioned herein are speculative in nature and may involve risk to principal and interest. Any prices or levels shown are either historical or purely indicative. This material does not take into account the particular investment objectives or financial circumstances, objectives or needs of any specific investor, and are not intended as recommendations of particular securities, investment products, or other financial products or strategies to particular clients. Securities, investment products, other financial products or strategies discussed herein may not be suitable for all investors. The recipient of this report must make its own independent decisions regarding any securities, investment products or other financial products mentioned herein.

The material should not be provided to any person in a jurisdiction where its provision or use would be contrary to local laws, rules or regulations. This material is not to be reproduced or redistributed to any other person or published in whole or in part for any purpose absent the written consent of Ironsides Macroeconomics.

© 2025 Ironsides Macroeconomics LLC.