2026 Q3 Market Review

The third quarter of 2026 was shaped by three forces: a re-escalation of the Iran conflict that reversed the prior quarter’s ceasefire, an historic Treasury bond selloff that pushed yields to multi-decade highs, and the Federal Reserve’s return to a tightening bias that culminated in its first rate hike since July 2023. U.S. equity benchmarks closed the quarter modestly higher at the index level, but that headline masked a sharp deterioration in market breadth, while fixed income delivered its worst quarterly performance of the current cycle.

The Main Drivers of Q3 2026 Market Dynamics

Renewed Middle East Escalation (July): The conditional U.S.–Iran ceasefire that lifted markets in Q2 did not hold. Iran began attacking vessels in the Strait of Hormuz, prompting broad U.S. retaliatory strikes, while Houthi militants advanced to Yemen’s Red Sea coast and intensified strikes on Saudi energy infrastructure. September negotiations at the UN produced no concrete resolution, with both sides reluctant to cede leverage first.

The Fed’s Hawkish Turn (August): Hawkish commentary from Chair Warsh and other policymakers at the Jackson Hole symposium confirmed that the central bank’s tightening bias, first signaled at Warsh’s inaugural June FOMC meeting, was solidifying amid persistently above-target inflation. An August announcement expanding the Treasury’s long-end buyback program did little to slow the climb in yields.

Tightening Resumes, the Bond Market Breaks (September): The Fed hiked rates 25 basis points in September — its first increase since July 2023. The move compounded an already-historic bond selloff: the 10-year Treasury yield rose 87 basis points over the quarter, its sharpest quarterly increase since 1994, briefly surpassing its 2007 peak to reach its highest level since 2002. The 30-year yield climbed to 5.67%, a 24-year high. Equities, remarkably, held near record levels throughout.

International Central Bank Divergence

International central banks diverged further in the third quarter. The ECB held rates steady at 2.25% in July before raising its deposit rate 25 basis points to 2.50% in September — its second hike of the cycle — as the Iran conflict’s inflationary pass-through pushed eurozone inflation to 3.3% in August, a three-year high. The Bank of Japan followed a similar path, holding at 1.00% in July before raising 25 basis points to 1.25% in September, a 31-year high, amid accelerating inflation and an historically weak yen that prompted joint BoJ-U.S. currency intervention. The Bank of England, by contrast, held its benchmark rate at 3.75% throughout the quarter despite UK CPI inflation rising to 3.1% in August; the Committee’s 6-3 vote in September, with three members favoring an immediate hike to 4.00%, points to a widely expected move at the November meeting.

U.S. Equity Markets: Narrow Leadership, Fraying Breadth¹

Headline index returns told only part of the story. The Dow fell 2.70%, while the S&P 500 (+2.03%) and Nasdaq Composite (+2.47%) posted modest gains after a strong second quarter. Beneath the surface, breadth deteriorated sharply: the equal-weighted S&P 500 fell 2.23%, lagging the cap-weighted index by 426 basis points — the index’s largest quarterly outperformance over its equal-weight counterpart since the second quarter of 2025. By quarter-end, only about 25% of index constituents traded above their 50-day moving average, down from roughly 64% on June 30, and the equal-weight index was on pace for its seventh consecutive weekly decline. To further illustrate the extreme concentration of performance within the S&P 500 during the quarter, four stocks: MSFT, NVDA, AAPL and META — added about 300 points to the gain in the S&P index, representing more than 200% of the index’s total gain for the quarter, while the other 496 stocks subtracted approximately 150 points. Or as the noted philosopher, Bobby McFerrin² once said:

Here’s a little song I wrote
You might want to sing it note for note
Don’t worry
Be happy
In every life we have some trouble
But when you worry you make it double
Don’t worry
Be happy, don’t worry, be happy now

The AI trade itself paused for breath. The semiconductor-heavy SOX fell 11.4% after gaining more than 230% across the prior six quarters, while software staged a comeback — the IGV software ETF rose 17.5% as fears of SaaS disruption from AI eased. Mega-cap performance was mixed but mostly positive: Microsoft gained 37.5%, and Meta rose 28.7%, helped late in the quarter by the launch of its Muse AI agent. The pause in AI enthusiasm was driven less by fundamentals than by a more guarded market conversation: revelations about questionable AI model behavior prompted calls to slow frontier-model development (which President Trump resisted, wary of ceding ground to China), alongside renewed skepticism about AI monetization, open-source competition, circular financing structures, and local backlash against data center construction.

Sector performance reflected the quarter’s cross-currents:

• Outperformers: Energy (+16.48%), aided by crude’s rebound amid the unresolved Iran war and particular strength in refiners; Technology (+7.06%); Healthcare (+6.03%); Communication Services (+3.48%).
• Underperformers: Utilities (−13.03%), Industrials (−9.91%), Real Estate (−6.32%), Consumer Discretionary (−5.24%), Materials (−3.19%), Consumer Staples (−2.43%), and Financials (−0.48%) — a group whose common thread was sensitivity to the quarter’s surge in long-term rates.

Earnings remained a genuine bright spot through the cross-currents. Second-quarter S&P 500 earnings grew nearly 53% year-over-year — still above 30% even excluding Alphabet and Meta — with 86% of constituents beating estimates, the highest share since the second quarter of 2021, on blended revenue growth of nearly 16%. Third-quarter earnings growth is currently tracking above 29%, which would mark a third consecutive quarter above 25%.

The labor market picture darkened notably right at the quarter’s edge. The September payroll report — released just after the quarter closed — both missed badly (+29,000 versus a roughly 84,000–100,000 consensus) and revised the prior two months sharply lower: July now shows a loss of 10,000 jobs (previously reported as a gain of 21,000), and August was cut to +133,000 (from +162,000) — a combined 60,000-job downward revision. The unemployment rate rose to 4.2%, driven by more workers entering the labor force rather than renewed layoffs, while wage growth slowed to roughly 3.0% year-over-year, its weakest pace in about five years; the report immediately pared back odds of an October Fed hike. Core PCE inflation told a similarly revised story: the BEA’s annual methodology update, released alongside the August report on September 30, changed how portfolio management, legal services, and computer software prices are calculated — applied retroactively to 2021 — cutting the reported year-over-year core PCE rate to 3.0% from what would otherwise have been 3.3%. On a three-month annualized basis, core PCE is now running close to the Fed’s 2% target, even as the twelve-month rate remains above it for a 66th consecutive month. Consumer sentiment stayed depressed despite resilient actual spending.

Small-Caps and the Cost of Duration

Small-caps bore the brunt of the quarter’s rate shock. The Russell 2000 fell 7.52%, making it one of the clearest casualties of the Treasury selloff: smaller companies carry more floating-rate and near-term refinancing exposure, and rate-sensitive small-cap names were among the market’s weakest performers alongside the most-shorted stocks. This marks a sharp reversal from the second quarter, when small- and micro-caps led every major size cohort.

International Developed and Emerging Markets: Early Gains Reversed

International equities followed a similar arc to the U.S. in reverse order — strength early, weakness late. The MSCI EAFE Index, supported earlier in the quarter by solid earnings and a softer dollar, finished the quarter down 1.2% as the Iran escalation weighed on sentiment. Emerging markets followed the same trajectory: strong early gains gave way to declines concentrated in energy-importing countries as crude prices climbed, leaving the MSCI Emerging Markets Index down 0.2% for the quarter.

Fixed Income Markets: An Historic Selloff

By nearly every measure, this was fixed income’s most difficult quarter of the current cycle. Treasury yields rose sharply across the curve: the 2-year yield rose 73 basis points, its largest quarterly jump since the second quarter of 2023, while the 10-year and 30-year yields both reached levels not seen since 2002. One widely read market commentary put it bluntly: the U.S. bond market “just completed its worst quarter this century.” The Bloomberg U.S. Aggregate Bond Index’s year-to-date return turned negative for the first time in 2026, finishing September at −2.91% after declining 3.51% during the quarter.

Despite the move, credit markets showed only late-quarter signs of strain rather than early ones. Investment-grade spreads remained near multi-year tights for most of the quarter before widening modestly to roughly 80 basis points by quarter-end — the largest monthly widening since March. High-yield spreads moved further, climbing from about 260 bps at the end of Q2 to 312 basis points by September 30, a five-month high, as a record $38.5 billion of September high-yield issuance began to test the market’s capacity to absorb supply. A contributing, less-discussed factor: the unwinding of the yen carry trade, which strategist Ed Yardeni and others have tied to a broader return of fiscal discipline-minded “bond vigilantes” as central-bank-driven demand for government debt fades globally.

Commodities and Currencies

Energy dominated the commodity complex, consistent with the quarter’s geopolitical narrative. WTI crude rose 30.1%, with diesel surging 45.2% — outperforming gasoline’s 12.8% gain amid tight refinery capacity heading into the U.S. harvest and heating seasons. Precious metals advanced more modestly after Q2’s slide, with gold up 3.7% and silver up 1.6%. Industrial metals also firmed: copper gained roughly 7.9% during the quarter, touching a record high above $14,800 per metric ton in September, supported by tight physical markets alongside structural AI- and grid-related demand. The dollar index was little changed (+0.3%), as dollar strength against the euro was offset by weakness against the yen amid joint BoJ-U.S. intervention.

Digital Assets: A Sharp Reversal From Q2

Crypto staged one of its strongest reversals in years. Bitcoin rose roughly 42.7% during the quarter — its best third-quarter performance since 2017 — after a difficult first half of the year. The move was not a steady climb: prices were flat through July and much of August before accelerating sharply from late August onward. Spot Bitcoin ETFs flipped from roughly $5 billion in net outflows in Q2 to $6.34 billion in net inflows in Q3, their strongest quarter of the year, while Ethereum ETFs pulled in a further $3.05 billion as Ethereum itself outperformed Bitcoin, rising roughly 71% on the quarter; Solana gained a further 61%, broadening the rally beyond the two largest tokens.

Private Assets — Equity / Credit / Real Estate

Private markets remained defined by a widening gap between paper gains and realized liquidity. AI-related financings now account for nearly 80% of trailing twelve-month venture deal value, and while exit value hit a record — led by SpaceX’s IPO — fund distributions relative to net asset value remain near 7.9%, far below the 14.5% long-run average, leaving the asset class with what one industry analysis termed a “liquidity paradox.” OpenAI’s own IPO remained delayed through year end even as Anthropic’s valuation approached $2 trillion following a large corporate funding round, underscoring how selective the reopening IPO window has become. Tech-focused private equity fundraising showed signs of life — PSG Equity and Francisco Partners alone accounted for more than 80% of the $31.3 billion raised by tech-focused funds in the quarter — though broader PE exit activity stayed constrained by the same geopolitical and valuation-gap pressures that limited dealmaking in the first half of the year. Commercial real estate conditions were little changed from Q2: elevated borrowing costs continued to limit new development, while constrained supply supported existing assets.

That is the data. What follows is where we think it might lead.

Beneath these headline numbers sits a structural question we have been tracking for several quarters now, and one we think deserves more attention than it is currently getting from either Washington or Wall Street: what happens to the labor market — and to the entitlement programs that depend on it — as artificial intelligence begins to do the jobs that used to train the next generation of workers.

Who’s Actually Losing Their Job to the Robots?

As you can see in the data compiled so far this year, the gap has become difficult to explain away as noise. Recent college graduate unemployment (ages 22–27) is running at roughly 5.6% to 5.8%, versus a national unemployment rate of only 4.2% to 4.3% — a gap that has widened by roughly 1.6 percentage points over the past three years. Underemployment among that same cohort — meaning a graduate working in a role that does not require the degree they paid for — sits near 41% to 42.5%. Entry-level hiring at the firms adopting generative AI most aggressively has fallen by as much as 80% per quarter, and new-graduate hires now make up roughly 7% of total hiring at large technology firms, down sharply from prior years. Goldman Sachs currently estimates AI is already reducing U.S. employment by approximately 16,000 jobs per month. The pattern is not confined to the United States: UK youth unemployment reached 16.4% this spring, up from 14.2% a year earlier.

Chart 1

The Widening Gap: Recent Graduates vs. the National Rate

 

Source: TPW analysis; recent-graduate and national unemployment figures as cited above

What’s notable is that employers are not simply deleting entry-level postings — they are redefining them. A pattern researchers call “seniorization” has emerged alongside the hiring decline, in which nominally junior roles are quietly rewritten to demand the judgment and stakeholder management of a mid-career employee. The entry-level rung of the career ladder is not disappearing. It is being raised out of reach.

Chart 2

Entry-Level Hiring at the Most Aggressive AI Adopters

Source: SignalFire hiring data, as cited across 2026 labor-market analyses

Without question, this data invites a simple story, AI is taking entry-level jobs, and we would be doing you a disservice if we let it stand unchallenged. One Yale-affiliated analysis found entry-level headcount at the heaviest AI spenders actually grew over the two years examined, not shrank. A separate study tied to Federal Reserve Bank of New York researchers argues the hiring slowdown is better explained by a frozen general labor market and the well-worn “job ladder” effect young workers experience in every downturn, AI or not. Even the executives building this technology cannot agree with each other: one chief executive has warned of unemployment reaching 10% to 20% within five years, while another has more recently downplayed the odds of any imminent, broad-based disruption. The problem, as I see it, is that the causal question is genuinely unresolved — and we think it is worth saying so plainly rather than reaching for whichever version of the story is more exciting.

As we discussed in a previous letter, the arithmetic behind the Social Security trust fund’s early-2030s depletion date was already unforgiving before any of this. What is new is the mechanism by which AI adoption could make it worse. Social Security and Medicare are funded by payroll taxes on wage income; if AI adoption narrows the base of FICA-paying wage earners even modestly — by displacing exactly the entry-level and mid-career roles discussed above — the trust fund’s arithmetic gets harder, not easier, at precisely the moment it can least afford it.

Four Ways to Tax a Robot, and Zero Laws to Show for It

Washington has begun to notice, but the debate over what to do about it has already produced more analytical clarity than the underlying labor data has. Serious proposals split along a basic design axis: should the tax fall on the employer, triggered when an actual job is lost to AI, or on the AI vendor, triggered by how much of the technology gets sold or used? Neither approach requires anything as dramatic as declaring an AI agent a legal “worker” — both are simply choices about where to place the incidence of an equivalent tax obligation, and that choice matters enormously for who actually pays and how easily the tax gets dodged.

By way of example, four distinct designs are currently in circulation, and the differences are not cosmetic. An employer-side, displacement-triggered tax — the design South Korea has filed — targets actual job losses directly but requires proving a layoff was caused by AI rather than by ordinary business conditions, which is exactly the causation dispute we described above. A vendor-side, usage-triggered tax — the design behind a bill in Congress — is far easier to administer, since it simply taxes token sales and product revenue, but it misses any company that builds its own AI in-house rather than buying it from a vendor, and its cost is likely to be passed straight through to customers. A capital-based tax on corporate income and capital gains generally — the design OpenAI itself has proposed — is diffuse, politically easy to endorse, and falls on the entire economy’s capital income rather than on the specific firms making automation decisions. And an academic “imputed worker” model, not yet reflected in any filed legislation, would impute a hypothetical employee onto a company based on its industry’s historical profit-per-worker ratio and tax the excess as that ratio rises with automation — the design closest to a literal digital-worker payroll tax.

Chart 3

Four Current Designs for Taxing Automated Labor, by Point of Incidence

Source: TPW analysis of filed legislation and published proposals, as cited above

As you can see, none of this is hypothetical anymore. H.R. 10044, the AI Tax and Work Protection Act, was introduced by Rep. Greg Casar with two co-sponsors on August 6th; it would levy AI companies on token sales and product revenue, with the rate auto-escalating once national unemployment crosses 5%, funding a jobs program with the proceeds. The bill was referred to two House committees currently controlled by members who have not indicated support, and it has not yet advanced. South Korea’s bills, filed eight days later on August 14th, take the opposite design — charging employers directly for AI-driven layoffs — explicitly to close the gap critics identified in the Casar bill: a company running its own model in-house, rather than buying tokens from a vendor, would escape a pure usage tax entirely. It is worth noting that no version of a robot tax, in any design, has ever become law anywhere in the world — not South Korea’s bill, not the Casar bill, and not Bill Gates’s original 2017 proposal for a “robot tax”, which the European Parliament rejected that same year on similar grounds. Gates proposed a robot tax to slow the pace of automation and fund social safety nets. Gates argued that if a robot replaces a human worker, the robot should be taxed at a level equivalent to the income and payroll taxes that the human would have generated.

As it happens, Gates himself weighed back into this debate recently, and the record is worth updating. Nearly a decade after that original proposal drew mostly ridicule, he wrote that he remains “a big proponent” of taxing AI — specifically both AI tokens and robots — and pointed to a structural bias already built into the tax code that none of the four designs above directly address: hire a person and the payroll tax bill follows immediately; buy a robot and it can typically be written off in full as a business expense in the same year. That is closer to a fifth design than a restatement of the first — not a pure employer-side displacement tax or a pure vendor-side usage tax, but a correction to a tax code that already, quietly, subsidizes the machine over the person.

Which brings us back to OpenAI’s own proposal, and it deserves more scrutiny than a passing mention. The document — a thirteen-page policy paper titled “Industrial Policy for the Intelligence Age,” published April 6th — devotes exactly one paragraph to automated labor taxation, and that paragraph specifies no rate, no defined tax base, and no point of incidence. Compared against H.R. 10044’s named tax base and trigger, or South Korea’s named taxable event, OpenAI’s proposal commits the company to nothing that would appear on its own income statement. That is not an incidental gap; it is the central analytical fact about the document. A broad capital-based tax spreads cost across the entire US economy’s corporate and investment income, while a targeted usage tax of the kind the Casar bill proposes would fall disproportionately on OpenAI’s own revenue — and OpenAI’s own document endorses the former, not the latter. We are not naive enough to believe that is a coincidence. The timing is relevant context, not proof of bad faith: the paper arrived as OpenAI approached an early Fall 2026 IPO (since postponed) at a nearly $1+ Trillion valuation, shortly after converting from a nonprofit to a for-profit company — and Sam Altman has said on the record that positioning the company as a proposer of solutions is also a strategy to shape regulation before regulation shapes the company, which is about as direct an acknowledgment of that dynamic as any of the firm’s critics could have asked for.

We are not the only ones drawing that inference, either. In the same essay, Gates made nearly the identical point about the industry more broadly, writing that AI companies are welcome to propose solutions to the problems their own technology creates, but “we should not expect them to lead the charge,” since deciding these questions is not a technologist’s role in a democratic society. Coming from someone with his own financial ties to the sector, that is a more pointed admission than it might first appear.

Put simply: everyone agrees a problem is coming, and no one agrees who should pay for it.

By way of example, this is where the story loops back into something we have written about at length in prior letters: interest rates. A tax that merely delays an AI-driven return, rather than eliminating it, extends the duration of that cash flow stream — and a longer-duration cash flow is mechanically more sensitive to today’s higher-for-longer real rate environment than it would have been a decade ago. A tax that captures a permanent share of the return, by contrast, is simply a haircut to terminal value, independent of where rates sit. Which effect prevails depends entirely on which policy design, if any, eventually passes. We got a preview of the market’s sensitivity to this narrative in mid-August, when robotics and AI-infrastructure stocks sold off sharply alongside a nineteen-year high in Treasury yields — the same rate-sensitivity mechanism we have discussed since 2022, now showing up in a new corner of the market.

Chart 4

30-Year Treasury Yield: The Climb Didn’t Stop in August

Source: Factset, Trading Economics, and U.S. Treasury daily par yield curve data, Aug. 5–Oct. 1, 2026

As you can see, that August high was not the peak. The thirty-year yield pushed through it again in September and stood at 5.61% as of early October — roughly twenty-seven basis points above the level we described as a nineteen-year high just weeks earlier. Whatever duration-extension effect this policy debate eventually adds, it is layering onto a rate environment that is still moving in one direction.

The politics of this will not stay theoretical for long. 2028 will be the first U.S. presidential election in which the Social Security depletion date falls inside the winner’s own term, which raises the cost of a vague answer considerably. The coalition lines on this issue do not sort neatly by party — populist and establishment factions within both parties are already talking past each other on it — and we would not be surprised if it becomes a genuine campaign issue well before the 2028 primaries begin in earnest.

As an Intellectual Exercise: One Path Through the Next Five Years

Everything above describes what is already measurable. What follows is not that — it is a constructed exercise, not a forecast, and we want to be explicit about the distinction before offering it. Imagine standing in January 2032, looking back at how this played out, if the current trajectory ran roughly down the middle of expert opinion rather than veering sharply toward either the optimistic or the alarming end.

Chart 5

One Path Through the Next Five Years (Illustrative Scenario, Not a Forecast)

Source: TPW analysis, built on published AI capability-timeline forecasts and the mechanisms described above; illustrative scenario, not a forecast

2027 would be the year agentic AI crosses from halting and expensive to the default way enterprise software gets built, and the labor effect that was narrow and entry-level-specific in 2026 would climb the experience ladder — the same firms that restructured junior roles around “seniorization” now restructuring the mid-level roles those juniors would have grown into. A handful of states would pass workforce-impact disclosure laws and pilot local automation taxes: genuine experiments, not a coordinated national policy.

By 2028, something like two to three hours of daily knowledge-work would be automated, a scale large enough to show up in household income data rather than only in occupational surveys — and, as we discussed above, an issue neither party could afford to leave unaddressed in a presidential race.

2029, on the median of expert timelines, is when frontier labs’ own capability estimates cluster for what researchers call a “superhuman coder” milestone — a system capable enough that a lab would rather retire its engineering team than keep training it. Whether or not that specific threshold is crossed on schedule, this is plausibly the year the divergence between AI-exposed and AI-insulated occupations becomes visible in the aggregate data, not just in the 22–27 age cohort we cited above.

2030 is where the paths diverge nationally rather than converge: one industry estimate puts the annual economic value created by AI agents and robots by this point near $2.9 trillion — large enough that how it gets distributed becomes a first-order policy question, and the U.S., the EU, and China would plausibly answer it in three genuinely different ways.

By 2031, the Social Security financing question we raised earlier in this letter would no longer be an abstraction: a payroll tax base narrowed by several years of labor substitution would be meeting the trust fund’s own arithmetic at the same time, rather than as two separate problems arriving in sequence.

And by January 2032, the honest picture — in this exercise — is neither triumphant nor catastrophic, but unresolved: some sectors settled toward augmentation, others toward substitution, split roughly along the lines already visible today, with the technology strengthening work that requires physical presence, trust, or genuinely novel judgment, and eroding work that does not.

We want to be realistic about three ways this could go differently. Capability progress could simply plateau before any “superhuman coder” milestone arrives — a real possibility that a number of credible researchers argue for today. Policy could work better than this exercise assumes, through retraining infrastructure or fiscal tools not yet on the table. Or the whole path could move faster and land harder than the median case above, in line with the more aggressive end of the capability forecasts and the more dire unemployment warnings already being made public. This is one plausible route through genuine uncertainty, not a prediction of which route we will actually take.

The bottom line is that none of this changes what we do with your money today. The causal debate over AI and youth unemployment is unresolved, the policy response is unenacted, and the market’s reaction has so far been contained to the specific equities most exposed to the narrative. But it is exactly the kind of slow-moving, multi-year structural risk that a diversified portfolio is built to withstand without requiring us to correctly predict its timing or its resolution. We see no reason to alter our portfolio positions at this time, and we will continue to track each link in this chain as it develops.

“We tend to overestimate the effect of a technology in the short run and underestimate the effect in the long run.”³

We at Twelve Points Wealth hope you and your families are well and enjoy the upcoming Holiday Season. Please call or email if you have any questions.

Steve

Steve Bruno, CFA

Chief Investment Officer

October 5, 2026

---

Notes

1. Data Sources: All quantitative financial metrics, index performance figures, and sector attribution are derived from FactSet Research Systems Inc., Bloomberg Index Services Limited, MSCI Inc., ICE/BofA, CoinGlass, and PitchBook, as of September 30, 2026, unless otherwise noted.
2. McFerrin, Bobby. “Don’t Worry, Be Happy.” Track 1 on Simple Pleasures. EMI-Manhattan Records, 1988, vinyl.
3. Attributed to Roy Amara, futurist and former president of the Institute for the Future — now generally known as “Amara’s Law.”

Important Disclosure Information

Please remember that past performance is no guarantee of future results. Different types of investments involve varying degrees of risk, and there can be no assurance that the future performance of any specific investment, investment strategy, or product (including the investments and/or investment strategies recommended or undertaken by Twelve Points Wealth Management (“Twelve Points”), or any non-investment related content, made reference to directly or indirectly in this commentary will be profitable, equal any corresponding indicated historical performance level(s), be suitable for your portfolio or individual situation, or prove successful. Due to various factors, including changing market conditions and/or applicable laws, the content may no longer be reflective of current opinions or positions. Moreover, no portion of the foregoing content serves as the receipt of, or a substitute for, personalized investment advice from Twelve Points. Neither Twelve Points’s investment adviser registration status, nor any amount of prior experience or success, should be construed that a certain level of results or satisfaction will be achieved if Twelve Points is engaged, or continues to be engaged, to provide investment advisory services. Twelve Points is neither a law firm, nor a certified public accounting firm, and no portion of the commentary content should be construed as legal or accounting advice. A copy of the Twelve Points’s current written disclosure Brochure and Form CRS discussing our advisory services and fees continues to remain available upon request or at www.twelvepoints.com. Please Remember: If you are a Twelve Points client, please contact Twelve Points, in writing, if there are any changes in your personal/financial situation or investment objectives for the purpose of reviewing/evaluating/revising our previous recommendations and/or services, or if you would like to impose, add, or to modify any reasonable restrictions to our investment advisory services. Unless, and until, you notify us, in writing, to the contrary, we shall continue to provide services as we do currently. Please Also Remember to advise us if you have not been receiving account statements (at least quarterly) from the account custodian.

Historical performance results for investment indices, benchmarks, and/or categories have been provided for general informational/comparison purposes only, and generally do not reflect the deduction of transaction and/or custodial charges, the deduction of an investment management fee, nor the impact of taxes, the incurrence of which would have the effect of decreasing historical performance results. It should not be assumed that your Twelve Points account holdings correspond directly to any comparative indices or categories. Please Also Note: (1) comparative benchmarks/indices may be more or less volatile than your Twelve Points accounts; and, (2) a description of each comparative benchmark/index is available upon request.

*Please Note: Ranking Limitations. Neither rankings nor recognitions by unaffiliated rating services, publications, media, or other organizations, nor the achievement of any professional designation, certification, degree, or license, membership in any professional organization, or any amount of prior experience or success, should be construed by a client or prospective client as a guarantee that the client will experience a certain level of results if the investment professional or the investment professional’s firm is engaged, or continues to be engaged, to provide investment advisory services. A fee was not paid by either the investment professional or the investment professional’s firm to receive the ranking. The ranking is based upon specific criteria and methodology (see ranking criteria/methodology). No ranking or recognition should be construed as an endorsement by any past or current client of the investment professional or the investment professional’s firm. ANY QUESTIONS: Dave Clayman, Twelve Points’s Chief Compliance Officer remains available to address any questions regarding rankings and/or recognitions, including the criteria used for any reflected ranking.

Disclaimer: Historical data is not a guarantee that any of the events described will occur or that any strategy will be successful. Past performance is not indicative of future results. Returns and data cited above are from various sources including FactSet, Bloomberg, Russell Associates, S&P Dow Jones, MSCI Inc., The St. Louis Federal Reserve, and FactSet, Inc. The content is developed from sources believed to be providing accurate information.

The information in this material is not intended as tax or legal advice. Please consult legal or tax professionals for specific information regarding your individual situation. The opinions expressed and material provided are for general information and should not be considered a solicitation for the purchase or sale of any security. Investing involves risks, including possible loss of principal. Please consider the investment objectives, risks, charges, and expenses of any security carefully before investing.

Twelve Points Wealth Management, LLC is an investment adviser located in Concord, Massachusetts. Twelve Points Wealth Management, LLC is registered with the Securities and Exchange Commission (SEC). Registration of an investment adviser does not imply any specific level of skill or training and does not constitute an endorsement of the firm by the Commission. Twelve Points Wealth Management, LLC only transacts business in states in which it is properly registered or is excluded or exempted from registration.